Merge pull request #4587 from ddasilva/meltingwax/4571

BUG: Explicitly reject nan values for p in binomial(n, p).

3 files changed, 1339 insertions, 1259 deletions

diff --git a/numpy/random/mtrand/mtrand.c b/numpy/random/mtrand/mtrand.c
index eac1f34d7..31475d986 100644
--- a/numpy/random/mtrand/mtrand.c
+++ b/numpy/random/mtrand/mtrand.c
@@ -1,4 +1,4 @@
-/* Generated by Cython 0.19.2 on Thu Mar 13 15:49:44 2014 */
+/* Generated by Cython 0.19 on Fri Apr  4 21:57:11 2014 */
 
 #define PY_SSIZE_T_CLEAN
 #ifndef CYTHON_USE_PYLONG_INTERNALS
@@ -264,7 +264,7 @@
 #ifndef CYTHON_RESTRICT
   #if defined(__GNUC__)
     #define CYTHON_RESTRICT __restrict__
-  #elif defined(_MSC_VER) && _MSC_VER >= 1400
+  #elif defined(_MSC_VER)
     #define CYTHON_RESTRICT __restrict
   #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L
     #define CYTHON_RESTRICT restrict
@@ -639,14 +639,6 @@ struct __pyx_obj_6mtrand_RandomState {
   #define __Pyx_XGOTREF(r)
   #define __Pyx_XGIVEREF(r)
 #endif /* CYTHON_REFNANNY */
-#define __Pyx_XDECREF_SET(r, v) do {                            \
-        PyObject *tmp = (PyObject *) r;                         \
-        r = v; __Pyx_XDECREF(tmp);                              \
-    } while (0)
-#define __Pyx_DECREF_SET(r, v) do {                             \
-        PyObject *tmp = (PyObject *) r;                         \
-        r = v; __Pyx_DECREF(tmp);                               \
-    } while (0)
 #define __Pyx_CLEAR(r)    do { PyObject* tmp = ((PyObject*)(r)); r = NULL; __Pyx_DECREF(tmp);} while(0)
 #define __Pyx_XCLEAR(r)   do { if((r) != NULL) {PyObject* tmp = ((PyObject*)(r)); r = NULL; __Pyx_DECREF(tmp);}} while(0)
 
@@ -854,11 +846,11 @@ static int __Pyx_InitStrings(__Pyx_StringTabEntry *t); /*proto*/
 /* Module declarations from 'numpy' */
 
 /* Module declarations from 'mtrand' */
+static PyTypeObject *__pyx_ptype_6mtrand_dtype = 0;
 static PyTypeObject *__pyx_ptype_6mtrand_ndarray = 0;
+static PyTypeObject *__pyx_ptype_6mtrand_flatiter = 0;
 static PyTypeObject *__pyx_ptype_6mtrand_broadcast = 0;
 static PyTypeObject *__pyx_ptype_6mtrand_RandomState = 0;
-static PyTypeObject *__pyx_ptype_6mtrand_flatiter = 0;
-static PyTypeObject *__pyx_ptype_6mtrand_dtype = 0;
 static PyObject *__pyx_f_6mtrand_cont0_array(rk_state *, __pyx_t_6mtrand_rk_cont0, PyObject *); /*proto*/
 static PyObject *__pyx_f_6mtrand_cont1_array_sc(rk_state *, __pyx_t_6mtrand_rk_cont1, PyObject *, double); /*proto*/
 static PyObject *__pyx_f_6mtrand_cont1_array(rk_state *, __pyx_t_6mtrand_rk_cont1, PyObject *, PyArrayObject *); /*proto*/
@@ -983,114 +975,115 @@ static char __pyx_k_130[] = "left == right";
 static char __pyx_k_135[] = "n < 0";
 static char __pyx_k_137[] = "p < 0";
 static char __pyx_k_139[] = "p > 1";
-static char __pyx_k_144[] = "n <= 0";
-static char __pyx_k_152[] = "lam < 0";
-static char __pyx_k_154[] = "lam value too large";
-static char __pyx_k_157[] = "lam value too large.";
-static char __pyx_k_159[] = "a <= 1.0";
-static char __pyx_k_162[] = "p < 0.0";
-static char __pyx_k_164[] = "p > 1.0";
-static char __pyx_k_168[] = "ngood < 0";
-static char __pyx_k_170[] = "nbad < 0";
-static char __pyx_k_172[] = "nsample < 1";
-static char __pyx_k_174[] = "ngood + nbad < nsample";
-static char __pyx_k_180[] = "p <= 0.0";
-static char __pyx_k_182[] = "p >= 1.0";
-static char __pyx_k_186[] = "numpy.dual";
-static char __pyx_k_187[] = "mean must be 1 dimensional";
-static char __pyx_k_189[] = "cov must be 2 dimensional and square";
-static char __pyx_k_191[] = "mean and cov must have same length";
-static char __pyx_k_194[] = "covariance is not positive-semidefinite.";
-static char __pyx_k_197[] = "sum(pvals[:-1]) > 1.0";
-static char __pyx_k_201[] = "/home/charris/Workspace/numpy.git/numpy/random/mtrand/mtrand.pyx";
-static char __pyx_k_204[] = "standard_exponential";
-static char __pyx_k_205[] = "noncentral_chisquare";
-static char __pyx_k_206[] = "RandomState.random_sample (line 731)";
-static char __pyx_k_207[] = "\n        random_sample(size=None)\n\n        Return random floats in the half-open interval [0.0, 1.0).\n\n        Results are from the \"continuous uniform\" distribution over the\n        stated interval.  To sample :math:`Unif[a, b), b > a` multiply\n        the output of `random_sample` by `(b-a)` and add `a`::\n\n          (b - a) * random_sample() + a\n\n        Parameters\n        ----------\n        size : int or tuple of ints, optional\n            Defines the shape of the returned array of random floats. If None\n            (the default), returns a single float.\n\n        Returns\n        -------\n        out : float or ndarray of floats\n            Array of random floats of shape `size` (unless ``size=None``, in which\n            case a single float is returned).\n\n        Examples\n        --------\n        >>> np.random.random_sample()\n        0.47108547995356098\n        >>> type(np.random.random_sample())\n        <type 'float'>\n        >>> np.random.random_sample((5,))\n        array([ 0.30220482,  0.86820401,  0.1654503 ,  0.11659149,  0.54323428])\n\n        Three-by-two array of random numbers from [-5, 0):\n\n        >>> 5 * np.random.random_sample((3, 2)) - 5\n        array([[-3.99149989, -0.52338984],\n               [-2.99091858, -0.79479508],\n               [-1.23204345, -1.75224494]])\n\n        ";
-static char __pyx_k_208[] = "RandomState.tomaxint (line 774)";
-static char __pyx_k_209[] = "\n        tomaxint(size=None)\n\n        Random integers between 0 and ``sys.maxint``, inclusive.\n\n        Return a sample of uniformly distributed random integers in the interval\n        [0, ``sys.maxint``].\n\n        Parameters\n        ----------\n        size : tuple of ints, int, optional\n            Shape of output.  If this is, for example, (m,n,k), m*n*k samples\n            are generated.  If no shape is specified, a single sample is\n            returned.\n\n        Returns\n        -------\n        out : ndarray\n            Drawn samples, with shape `size`.\n\n        See Also\n        --------\n        randint : Uniform sampling over a given half-open interval of integers.\n        random_integers : Uniform sampling over a given closed interval of\n            integers.\n\n        Examples\n        --------\n        >>> RS = np.random.mtrand.RandomState() # need a RandomState object\n        >>> RS.tomaxint((2,2,2))\n        array([[[1170048599, 1600360186],\n                [ 739731006, 1947757578]],\n               [[1871712945,  752307660],\n                [1601631370, 1479324245]]])\n        >>> import sys\n        >>> sys.maxint\n        2147483647\n        >>> RS.tomaxint((2,2,2)) < sys.maxint\n        array([[[ True,  True],\n                [ True,  True]],\n               [[ True,  True],\n                [ True,  True]]], dtype=bool)\n\n        ";
-static char __pyx_k_210[] = "RandomState.randint (line 821)";
-static char __pyx_k_211[] = "\n        randint(low, high=None, size=None)\n\n        Return random integers from `low` (inclusive) to `high` (exclusive).\n\n        Return random integers from the \"discrete uniform\" distribution in the\n        \"half-open\" interval [`low`, `high`). If `high` is None (the default),\n        then results are from [0, `low`).\n\n        Parameters\n        ----------\n        low : int\n            Lowest (signed) integer to be drawn from the distribution (unless\n            ``high=None``, in which case this parameter is the *highest* such\n            integer).\n        high : int, optional\n            If provided, one above the largest (signed) integer to be drawn\n            from the distribution (see above for behavior if ``high=None``).\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single int is\n            returned.\n\n        Returns\n        -------\n        out : int or ndarray of ints\n            `size`-shaped array of random integers from the appropriate\n            distribution, or a single such random int if `size` not provided.\n\n        See Also\n        --------\n        random.random_integers : similar to `randint`, only for the closed\n            interval [`low`, `high`], and 1 is the lowest value if `high` is\n            omitted. In particular, this other one is the one to use to generate\n            uniformly distributed discrete non-integers.\n\n        Examples\n        --------\n        >>> np.random.randint(2, size=10)\n        array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0])\n        >>> np.random.randint(1, size=10)\n        array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0])\n\n        Generate a 2 x 4 array of ints between 0 and 4, inclusive:\n\n        >>> np.random.randint(5, size=(2, 4))\n        array([[4, 0, 2, 1],\n               [3, 2, 2, 0]])\n\n        ";
-static char __pyx_k_212[] = "RandomState.bytes (line 901)";
-static char __pyx_k_213[] = "\n        bytes(length)\n\n        Return random bytes.\n\n        Parameters\n        ----------\n        length : int\n            Number of random bytes.\n\n        Returns\n        -------\n        out : str\n            String of length `length`.\n\n        Examples\n        --------\n        >>> np.random.bytes(10)\n        ' eh\\x85\\x022SZ\\xbf\\xa4' #random\n\n        ";
-static char __pyx_k_214[] = "RandomState.choice (line 929)";
-static char __pyx_k_215[] = "\n        choice(a, size=None, replace=True, p=None)\n\n        Generates a random sample from a given 1-D array\n\n                .. versionadded:: 1.7.0\n\n        Parameters\n        -----------\n        a : 1-D array-like or int\n            If an ndarray, a random sample is generated from its elements.\n            If an int, the random sample is generated as if a was np.arange(n)\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n        replace : boolean, optional\n            Whether the sample is with or without replacement\n        p : 1-D array-like, optional\n            The probabilities associated with each entry in a.\n            If not given the sample assumes a uniform distribtion over all\n            entries in a.\n\n        Returns\n        --------\n        samples : 1-D ndarray, shape (size,)\n            The generated random samples\n\n        Raises\n        -------\n        ValueError\n            If a is an int and less than zero, if a or p are not 1-dimensional,\n            if a is an array-like of size 0, if p is not a vector of\n            probabilities, if a and p have different lengths, or if\n            replace=False and the sample size is greater than the population\n            size\n\n        See Also\n        ---------\n        randint, shuffle, permutation\n\n        Examples\n        ---------\n        Generate a uniform random sample from np.arange(5) of size 3:\n\n        >>> np.random.choice(5, 3)\n        array([0, 3, 4])\n        >>> #This is equivalent to np.random.randint(0,5,3)\n\n        Generate a non-uniform random sample from np.arange(5) of size 3:\n\n        >>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0])\n        array([3, 3, 0])\n\n        Generate a uniform random sample from np.arange(5) of size 3 without\n        replacement:\n\n        >>> np.random.choice(5, 3, replace=False)\n        array([3,1,0])\n    ""    >>> #This is equivalent to np.random.permutation(np.arange(5))[:3]\n\n        Generate a non-uniform random sample from np.arange(5) of size\n        3 without replacement:\n\n        >>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0])\n        array([2, 3, 0])\n\n        Any of the above can be repeated with an arbitrary array-like\n        instead of just integers. For instance:\n\n        >>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher']\n        >>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3])\n        array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'],\n              dtype='|S11')\n\n        ";
-static char __pyx_k_216[] = "RandomState.uniform (line 1104)";
-static char __pyx_k_217[] = "\n        uniform(low=0.0, high=1.0, size=1)\n\n        Draw samples from a uniform distribution.\n\n        Samples are uniformly distributed over the half-open interval\n        ``[low, high)`` (includes low, but excludes high).  In other words,\n        any value within the given interval is equally likely to be drawn\n        by `uniform`.\n\n        Parameters\n        ----------\n        low : float, optional\n            Lower boundary of the output interval.  All values generated will be\n            greater than or equal to low.  The default value is 0.\n        high : float\n            Upper boundary of the output interval.  All values generated will be\n            less than high.  The default value is 1.0.\n        size : int or tuple of ints, optional\n            Shape of output.  If the given size is, for example, (m,n,k),\n            m*n*k samples are generated.  If no shape is specified, a single sample\n            is returned.\n\n        Returns\n        -------\n        out : ndarray\n            Drawn samples, with shape `size`.\n\n        See Also\n        --------\n        randint : Discrete uniform distribution, yielding integers.\n        random_integers : Discrete uniform distribution over the closed\n                          interval ``[low, high]``.\n        random_sample : Floats uniformly distributed over ``[0, 1)``.\n        random : Alias for `random_sample`.\n        rand : Convenience function that accepts dimensions as input, e.g.,\n               ``rand(2,2)`` would generate a 2-by-2 array of floats,\n               uniformly distributed over ``[0, 1)``.\n\n        Notes\n        -----\n        The probability density function of the uniform distribution is\n\n        .. math:: p(x) = \\frac{1}{b - a}\n\n        anywhere within the interval ``[a, b)``, and zero elsewhere.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> s = np.random.uniform(-1,0,1000)\n\n        All values are w""ithin the given interval:\n\n        >>> np.all(s >= -1)\n        True\n        >>> np.all(s < 0)\n        True\n\n        Display the histogram of the samples, along with the\n        probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 15, normed=True)\n        >>> plt.plot(bins, np.ones_like(bins), linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_218[] = "RandomState.rand (line 1191)";
-static char __pyx_k_219[] = "\n        rand(d0, d1, ..., dn)\n\n        Random values in a given shape.\n\n        Create an array of the given shape and propagate it with\n        random samples from a uniform distribution\n        over ``[0, 1)``.\n\n        Parameters\n        ----------\n        d0, d1, ..., dn : int, optional\n            The dimensions of the returned array, should all be positive.\n            If no argument is given a single Python float is returned.\n\n        Returns\n        -------\n        out : ndarray, shape ``(d0, d1, ..., dn)``\n            Random values.\n\n        See Also\n        --------\n        random\n\n        Notes\n        -----\n        This is a convenience function. If you want an interface that\n        takes a shape-tuple as the first argument, refer to\n        np.random.random_sample .\n\n        Examples\n        --------\n        >>> np.random.rand(3,2)\n        array([[ 0.14022471,  0.96360618],  #random\n               [ 0.37601032,  0.25528411],  #random\n               [ 0.49313049,  0.94909878]]) #random\n\n        ";
-static char __pyx_k_220[] = "RandomState.randn (line 1235)";
-static char __pyx_k_221[] = "\n        randn(d0, d1, ..., dn)\n\n        Return a sample (or samples) from the \"standard normal\" distribution.\n\n        If positive, int_like or int-convertible arguments are provided,\n        `randn` generates an array of shape ``(d0, d1, ..., dn)``, filled\n        with random floats sampled from a univariate \"normal\" (Gaussian)\n        distribution of mean 0 and variance 1 (if any of the :math:`d_i` are\n        floats, they are first converted to integers by truncation). A single\n        float randomly sampled from the distribution is returned if no\n        argument is provided.\n\n        This is a convenience function.  If you want an interface that takes a\n        tuple as the first argument, use `numpy.random.standard_normal` instead.\n\n        Parameters\n        ----------\n        d0, d1, ..., dn : int, optional\n            The dimensions of the returned array, should be all positive.\n            If no argument is given a single Python float is returned.\n\n        Returns\n        -------\n        Z : ndarray or float\n            A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from\n            the standard normal distribution, or a single such float if\n            no parameters were supplied.\n\n        See Also\n        --------\n        random.standard_normal : Similar, but takes a tuple as its argument.\n\n        Notes\n        -----\n        For random samples from :math:`N(\\mu, \\sigma^2)`, use:\n\n        ``sigma * np.random.randn(...) + mu``\n\n        Examples\n        --------\n        >>> np.random.randn()\n        2.1923875335537315 #random\n\n        Two-by-four array of samples from N(3, 6.25):\n\n        >>> 2.5 * np.random.randn(2, 4) + 3\n        array([[-4.49401501,  4.00950034, -1.81814867,  7.29718677],  #random\n               [ 0.39924804,  4.68456316,  4.99394529,  4.84057254]]) #random\n\n        ";
-static char __pyx_k_222[] = "RandomState.random_integers (line 1292)";
-static char __pyx_k_223[] = "\n        random_integers(low, high=None, size=None)\n\n        Return random integers between `low` and `high`, inclusive.\n\n        Return random integers from the \"discrete uniform\" distribution in the\n        closed interval [`low`, `high`].  If `high` is None (the default),\n        then results are from [1, `low`].\n\n        Parameters\n        ----------\n        low : int\n            Lowest (signed) integer to be drawn from the distribution (unless\n            ``high=None``, in which case this parameter is the *highest* such\n            integer).\n        high : int, optional\n            If provided, the largest (signed) integer to be drawn from the\n            distribution (see above for behavior if ``high=None``).\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single int is returned.\n\n        Returns\n        -------\n        out : int or ndarray of ints\n            `size`-shaped array of random integers from the appropriate\n            distribution, or a single such random int if `size` not provided.\n\n        See Also\n        --------\n        random.randint : Similar to `random_integers`, only for the half-open\n            interval [`low`, `high`), and 0 is the lowest value if `high` is\n            omitted.\n\n        Notes\n        -----\n        To sample from N evenly spaced floating-point numbers between a and b,\n        use::\n\n          a + (b - a) * (np.random.random_integers(N) - 1) / (N - 1.)\n\n        Examples\n        --------\n        >>> np.random.random_integers(5)\n        4\n        >>> type(np.random.random_integers(5))\n        <type 'int'>\n        >>> np.random.random_integers(5, size=(3.,2.))\n        array([[5, 4],\n               [3, 3],\n               [4, 5]])\n\n        Choose five random numbers from the set of five evenly-spaced\n        numbers between 0 and 2.5, inclusive (*i.e.*, from the set\n        :math:`{0, 5/8, 10/8, 15/8, 20/8}`):\n""\n        >>> 2.5 * (np.random.random_integers(5, size=(5,)) - 1) / 4.\n        array([ 0.625,  1.25 ,  0.625,  0.625,  2.5  ])\n\n        Roll two six sided dice 1000 times and sum the results:\n\n        >>> d1 = np.random.random_integers(1, 6, 1000)\n        >>> d2 = np.random.random_integers(1, 6, 1000)\n        >>> dsums = d1 + d2\n\n        Display results as a histogram:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(dsums, 11, normed=True)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_224[] = "RandomState.standard_normal (line 1370)";
-static char __pyx_k_225[] = "\n        standard_normal(size=None)\n\n        Returns samples from a Standard Normal distribution (mean=0, stdev=1).\n\n        Parameters\n        ----------\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        out : float or ndarray\n            Drawn samples.\n\n        Examples\n        --------\n        >>> s = np.random.standard_normal(8000)\n        >>> s\n        array([ 0.6888893 ,  0.78096262, -0.89086505, ...,  0.49876311, #random\n               -0.38672696, -0.4685006 ])                               #random\n        >>> s.shape\n        (8000,)\n        >>> s = np.random.standard_normal(size=(3, 4, 2))\n        >>> s.shape\n        (3, 4, 2)\n\n        ";
-static char __pyx_k_226[] = "RandomState.normal (line 1402)";
-static char __pyx_k_227[] = "\n        normal(loc=0.0, scale=1.0, size=None)\n\n        Draw random samples from a normal (Gaussian) distribution.\n\n        The probability density function of the normal distribution, first\n        derived by De Moivre and 200 years later by both Gauss and Laplace\n        independently [2]_, is often called the bell curve because of\n        its characteristic shape (see the example below).\n\n        The normal distributions occurs often in nature.  For example, it\n        describes the commonly occurring distribution of samples influenced\n        by a large number of tiny, random disturbances, each with its own\n        unique distribution [2]_.\n\n        Parameters\n        ----------\n        loc : float\n            Mean (\"centre\") of the distribution.\n        scale : float\n            Standard deviation (spread or \"width\") of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        See Also\n        --------\n        scipy.stats.distributions.norm : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Gaussian distribution is\n\n        .. math:: p(x) = \\frac{1}{\\sqrt{ 2 \\pi \\sigma^2 }}\n                         e^{ - \\frac{ (x - \\mu)^2 } {2 \\sigma^2} },\n\n        where :math:`\\mu` is the mean and :math:`\\sigma` the standard deviation.\n        The square of the standard deviation, :math:`\\sigma^2`, is called the\n        variance.\n\n        The function has its peak at the mean, and its \"spread\" increases with\n        the standard deviation (the function reaches 0.607 times its maximum at\n        :math:`x + \\sigma` and :math:`x - \\sigma` [2]_).  This implies that\n        `numpy.random.normal` is more likely to return samples lying close to the\n        mean, rather than those far away.\n""\n        References\n        ----------\n        .. [1] Wikipedia, \"Normal distribution\",\n               http://en.wikipedia.org/wiki/Normal_distribution\n        .. [2] P. R. Peebles Jr., \"Central Limit Theorem\" in \"Probability, Random\n               Variables and Random Signal Principles\", 4th ed., 2001,\n               pp. 51, 51, 125.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, sigma = 0, 0.1 # mean and standard deviation\n        >>> s = np.random.normal(mu, sigma, 1000)\n\n        Verify the mean and the variance:\n\n        >>> abs(mu - np.mean(s)) < 0.01\n        True\n\n        >>> abs(sigma - np.std(s, ddof=1)) < 0.01\n        True\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 30, normed=True)\n        >>> plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) *\n        ...                np.exp( - (bins - mu)**2 / (2 * sigma**2) ),\n        ...          linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_228[] = "RandomState.standard_exponential (line 1615)";
-static char __pyx_k_229[] = "\n        standard_exponential(size=None)\n\n        Draw samples from the standard exponential distribution.\n\n        `standard_exponential` is identical to the exponential distribution\n        with a scale parameter of 1.\n\n        Parameters\n        ----------\n        size : int or tuple of ints\n            Shape of the output.\n\n        Returns\n        -------\n        out : float or ndarray\n            Drawn samples.\n\n        Examples\n        --------\n        Output a 3x8000 array:\n\n        >>> n = np.random.standard_exponential((3, 8000))\n\n        ";
-static char __pyx_k_230[] = "RandomState.standard_gamma (line 1643)";
-static char __pyx_k_231[] = "\n        standard_gamma(shape, size=None)\n\n        Draw samples from a Standard Gamma distribution.\n\n        Samples are drawn from a Gamma distribution with specified parameters,\n        shape (sometimes designated \"k\") and scale=1.\n\n        Parameters\n        ----------\n        shape : float\n            Parameter, should be > 0.\n        size : int or tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The drawn samples.\n\n        See Also\n        --------\n        scipy.stats.distributions.gamma : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Gamma distribution is\n\n        .. math:: p(x) = x^{k-1}\\frac{e^{-x/\\theta}}{\\theta^k\\Gamma(k)},\n\n        where :math:`k` is the shape and :math:`\\theta` the scale,\n        and :math:`\\Gamma` is the Gamma function.\n\n        The Gamma distribution is often used to model the times to failure of\n        electronic components, and arises naturally in processes for which the\n        waiting times between Poisson distributed events are relevant.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Gamma Distribution.\" From MathWorld--A\n               Wolfram Web Resource.\n               http://mathworld.wolfram.com/GammaDistribution.html\n        .. [2] Wikipedia, \"Gamma-distribution\",\n               http://en.wikipedia.org/wiki/Gamma-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> shape, scale = 2., 1. # mean and width\n        >>> s = np.random.standard_gamma(shape, 1000000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt""\n        >>> import scipy.special as sps\n        >>> count, bins, ignored = plt.hist(s, 50, normed=True)\n        >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ \\\n        ...                       (sps.gamma(shape) * scale**shape))\n        >>> plt.plot(bins, y, linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_232[] = "RandomState.gamma (line 1725)";
-static char __pyx_k_233[] = "\n        gamma(shape, scale=1.0, size=None)\n\n        Draw samples from a Gamma distribution.\n\n        Samples are drawn from a Gamma distribution with specified parameters,\n        `shape` (sometimes designated \"k\") and `scale` (sometimes designated\n        \"theta\"), where both parameters are > 0.\n\n        Parameters\n        ----------\n        shape : scalar > 0\n            The shape of the gamma distribution.\n        scale : scalar > 0, optional\n            The scale of the gamma distribution.  Default is equal to 1.\n        size : shape_tuple, optional\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        out : ndarray, float\n            Returns one sample unless `size` parameter is specified.\n\n        See Also\n        --------\n        scipy.stats.distributions.gamma : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Gamma distribution is\n\n        .. math:: p(x) = x^{k-1}\\frac{e^{-x/\\theta}}{\\theta^k\\Gamma(k)},\n\n        where :math:`k` is the shape and :math:`\\theta` the scale,\n        and :math:`\\Gamma` is the Gamma function.\n\n        The Gamma distribution is often used to model the times to failure of\n        electronic components, and arises naturally in processes for which the\n        waiting times between Poisson distributed events are relevant.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Gamma Distribution.\" From MathWorld--A\n               Wolfram Web Resource.\n               http://mathworld.wolfram.com/GammaDistribution.html\n        .. [2] Wikipedia, \"Gamma-distribution\",\n               http://en.wikipedia.org/wiki/Gamma-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> shape, scale = 2.,"" 2. # mean and dispersion\n        >>> s = np.random.gamma(shape, scale, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> import scipy.special as sps\n        >>> count, bins, ignored = plt.hist(s, 50, normed=True)\n        >>> y = bins**(shape-1)*(np.exp(-bins/scale) /\n        ...                      (sps.gamma(shape)*scale**shape))\n        >>> plt.plot(bins, y, linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_234[] = "RandomState.f (line 1816)";
-static char __pyx_k_235[] = "\n        f(dfnum, dfden, size=None)\n\n        Draw samples from a F distribution.\n\n        Samples are drawn from an F distribution with specified parameters,\n        `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom\n        in denominator), where both parameters should be greater than zero.\n\n        The random variate of the F distribution (also known as the\n        Fisher distribution) is a continuous probability distribution\n        that arises in ANOVA tests, and is the ratio of two chi-square\n        variates.\n\n        Parameters\n        ----------\n        dfnum : float\n            Degrees of freedom in numerator. Should be greater than zero.\n        dfden : float\n            Degrees of freedom in denominator. Should be greater than zero.\n        size : {tuple, int}, optional\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``,\n            then ``m * n * k`` samples are drawn. By default only one sample\n            is returned.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n            Samples from the Fisher distribution.\n\n        See Also\n        --------\n        scipy.stats.distributions.f : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The F statistic is used to compare in-group variances to between-group\n        variances. Calculating the distribution depends on the sampling, and\n        so it is a function of the respective degrees of freedom in the\n        problem.  The variable `dfnum` is the number of samples minus one, the\n        between-groups degrees of freedom, while `dfden` is the within-groups\n        degrees of freedom, the sum of the number of samples in each group\n        minus the number of groups.\n\n        References\n        ----------\n        .. [1] Glantz, Stanton A. \"Primer of Biostatistics.\", McGraw-Hill,\n               Fifth Edition, 2002.""\n        .. [2] Wikipedia, \"F-distribution\",\n               http://en.wikipedia.org/wiki/F-distribution\n\n        Examples\n        --------\n        An example from Glantz[1], pp 47-40.\n        Two groups, children of diabetics (25 people) and children from people\n        without diabetes (25 controls). Fasting blood glucose was measured,\n        case group had a mean value of 86.1, controls had a mean value of\n        82.2. Standard deviations were 2.09 and 2.49 respectively. Are these\n        data consistent with the null hypothesis that the parents diabetic\n        status does not affect their children's blood glucose levels?\n        Calculating the F statistic from the data gives a value of 36.01.\n\n        Draw samples from the distribution:\n\n        >>> dfnum = 1. # between group degrees of freedom\n        >>> dfden = 48. # within groups degrees of freedom\n        >>> s = np.random.f(dfnum, dfden, 1000)\n\n        The lower bound for the top 1% of the samples is :\n\n        >>> sort(s)[-10]\n        7.61988120985\n\n        So there is about a 1% chance that the F statistic will exceed 7.62,\n        the measured value is 36, so the null hypothesis is rejected at the 1%\n        level.\n\n        ";
-static char __pyx_k_236[] = "RandomState.noncentral_f (line 1918)";
-static char __pyx_k_237[] = "\n        noncentral_f(dfnum, dfden, nonc, size=None)\n\n        Draw samples from the noncentral F distribution.\n\n        Samples are drawn from an F distribution with specified parameters,\n        `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of\n        freedom in denominator), where both parameters > 1.\n        `nonc` is the non-centrality parameter.\n\n        Parameters\n        ----------\n        dfnum : int\n            Parameter, should be > 1.\n        dfden : int\n            Parameter, should be > 1.\n        nonc : float\n            Parameter, should be >= 0.\n        size : int or tuple of ints\n            Output shape. If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : scalar or ndarray\n            Drawn samples.\n\n        Notes\n        -----\n        When calculating the power of an experiment (power = probability of\n        rejecting the null hypothesis when a specific alternative is true) the\n        non-central F statistic becomes important.  When the null hypothesis is\n        true, the F statistic follows a central F distribution. When the null\n        hypothesis is not true, then it follows a non-central F statistic.\n\n        References\n        ----------\n        Weisstein, Eric W. \"Noncentral F-Distribution.\" From MathWorld--A Wolfram\n        Web Resource.  http://mathworld.wolfram.com/NoncentralF-Distribution.html\n\n        Wikipedia, \"Noncentral F distribution\",\n        http://en.wikipedia.org/wiki/Noncentral_F-distribution\n\n        Examples\n        --------\n        In a study, testing for a specific alternative to the null hypothesis\n        requires use of the Noncentral F distribution. We need to calculate the\n        area in the tail of the distribution that exceeds the value of the F\n        distribution for the null hypothesis.  We'll plot the two probability\n        distributions for comp""arison.\n\n        >>> dfnum = 3 # between group deg of freedom\n        >>> dfden = 20 # within groups degrees of freedom\n        >>> nonc = 3.0\n        >>> nc_vals = np.random.noncentral_f(dfnum, dfden, nonc, 1000000)\n        >>> NF = np.histogram(nc_vals, bins=50, normed=True)\n        >>> c_vals = np.random.f(dfnum, dfden, 1000000)\n        >>> F = np.histogram(c_vals, bins=50, normed=True)\n        >>> plt.plot(F[1][1:], F[0])\n        >>> plt.plot(NF[1][1:], NF[0])\n        >>> plt.show()\n\n        ";
-static char __pyx_k_238[] = "RandomState.chisquare (line 2013)";
-static char __pyx_k_239[] = "\n        chisquare(df, size=None)\n\n        Draw samples from a chi-square distribution.\n\n        When `df` independent random variables, each with standard normal\n        distributions (mean 0, variance 1), are squared and summed, the\n        resulting distribution is chi-square (see Notes).  This distribution\n        is often used in hypothesis testing.\n\n        Parameters\n        ----------\n        df : int\n             Number of degrees of freedom.\n        size : tuple of ints, int, optional\n             Size of the returned array.  By default, a scalar is\n             returned.\n\n        Returns\n        -------\n        output : ndarray\n            Samples drawn from the distribution, packed in a `size`-shaped\n            array.\n\n        Raises\n        ------\n        ValueError\n            When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``)\n            is given.\n\n        Notes\n        -----\n        The variable obtained by summing the squares of `df` independent,\n        standard normally distributed random variables:\n\n        .. math:: Q = \\sum_{i=0}^{\\mathtt{df}} X^2_i\n\n        is chi-square distributed, denoted\n\n        .. math:: Q \\sim \\chi^2_k.\n\n        The probability density function of the chi-squared distribution is\n\n        .. math:: p(x) = \\frac{(1/2)^{k/2}}{\\Gamma(k/2)}\n                         x^{k/2 - 1} e^{-x/2},\n\n        where :math:`\\Gamma` is the gamma function,\n\n        .. math:: \\Gamma(x) = \\int_0^{-\\infty} t^{x - 1} e^{-t} dt.\n\n        References\n        ----------\n        `NIST/SEMATECH e-Handbook of Statistical Methods\n        <http://www.itl.nist.gov/div898/handbook/eda/section3/eda3666.htm>`_\n\n        Examples\n        --------\n        >>> np.random.chisquare(2,4)\n        array([ 1.89920014,  9.00867716,  3.13710533,  5.62318272])\n\n        ";
-static char __pyx_k_240[] = "RandomState.noncentral_chisquare (line 2091)";
-static char __pyx_k_241[] = "\n        noncentral_chisquare(df, nonc, size=None)\n\n        Draw samples from a noncentral chi-square distribution.\n\n        The noncentral :math:`\\chi^2` distribution is a generalisation of\n        the :math:`\\chi^2` distribution.\n\n        Parameters\n        ----------\n        df : int\n            Degrees of freedom, should be >= 1.\n        nonc : float\n            Non-centrality, should be > 0.\n        size : int or tuple of ints\n            Shape of the output.\n\n        Notes\n        -----\n        The probability density function for the noncentral Chi-square distribution\n        is\n\n        .. math:: P(x;df,nonc) = \\sum^{\\infty}_{i=0}\n                               \\frac{e^{-nonc/2}(nonc/2)^{i}}{i!}P_{Y_{df+2i}}(x),\n\n        where :math:`Y_{q}` is the Chi-square with q degrees of freedom.\n\n        In Delhi (2007), it is noted that the noncentral chi-square is useful in\n        bombing and coverage problems, the probability of killing the point target\n        given by the noncentral chi-squared distribution.\n\n        References\n        ----------\n        .. [1] Delhi, M.S. Holla, \"On a noncentral chi-square distribution in the\n               analysis of weapon systems effectiveness\", Metrika, Volume 15,\n               Number 1 / December, 1970.\n        .. [2] Wikipedia, \"Noncentral chi-square distribution\"\n               http://en.wikipedia.org/wiki/Noncentral_chi-square_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram\n\n        >>> import matplotlib.pyplot as plt\n        >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000),\n        ...                   bins=200, normed=True)\n        >>> plt.show()\n\n        Draw values from a noncentral chisquare with very small noncentrality,\n        and compare to a chisquare.\n\n        >>> plt.figure()\n        >>> values = plt.hist(np.random.noncentral_chisquare(3, .0000001, 100000),\n     ""   ...                   bins=np.arange(0., 25, .1), normed=True)\n        >>> values2 = plt.hist(np.random.chisquare(3, 100000),\n        ...                    bins=np.arange(0., 25, .1), normed=True)\n        >>> plt.plot(values[1][0:-1], values[0]-values2[0], 'ob')\n        >>> plt.show()\n\n        Demonstrate how large values of non-centrality lead to a more symmetric\n        distribution.\n\n        >>> plt.figure()\n        >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000),\n        ...                   bins=200, normed=True)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_242[] = "RandomState.standard_cauchy (line 2183)";
-static char __pyx_k_243[] = "\n        standard_cauchy(size=None)\n\n        Standard Cauchy distribution with mode = 0.\n\n        Also known as the Lorentz distribution.\n\n        Parameters\n        ----------\n        size : int or tuple of ints\n            Shape of the output.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The drawn samples.\n\n        Notes\n        -----\n        The probability density function for the full Cauchy distribution is\n\n        .. math:: P(x; x_0, \\gamma) = \\frac{1}{\\pi \\gamma \\bigl[ 1+\n                  (\\frac{x-x_0}{\\gamma})^2 \\bigr] }\n\n        and the Standard Cauchy distribution just sets :math:`x_0=0` and\n        :math:`\\gamma=1`\n\n        The Cauchy distribution arises in the solution to the driven harmonic\n        oscillator problem, and also describes spectral line broadening. It\n        also describes the distribution of values at which a line tilted at\n        a random angle will cut the x axis.\n\n        When studying hypothesis tests that assume normality, seeing how the\n        tests perform on data from a Cauchy distribution is a good indicator of\n        their sensitivity to a heavy-tailed distribution, since the Cauchy looks\n        very much like a Gaussian distribution, but with heavier tails.\n\n        References\n        ----------\n        .. [1] NIST/SEMATECH e-Handbook of Statistical Methods, \"Cauchy\n              Distribution\",\n              http://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm\n        .. [2] Weisstein, Eric W. \"Cauchy Distribution.\" From MathWorld--A\n              Wolfram Web Resource.\n              http://mathworld.wolfram.com/CauchyDistribution.html\n        .. [3] Wikipedia, \"Cauchy distribution\"\n              http://en.wikipedia.org/wiki/Cauchy_distribution\n\n        Examples\n        --------\n        Draw samples and plot the distribution:\n\n        >>> s = np.random.standard_cauchy(1000000)\n        >>> s = s[(s>-25) & (s<""25)]  # truncate distribution so it plots well\n        >>> plt.hist(s, bins=100)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_244[] = "RandomState.standard_t (line 2244)";
-static char __pyx_k_245[] = "\n        standard_t(df, size=None)\n\n        Standard Student's t distribution with df degrees of freedom.\n\n        A special case of the hyperbolic distribution.\n        As `df` gets large, the result resembles that of the standard normal\n        distribution (`standard_normal`).\n\n        Parameters\n        ----------\n        df : int\n            Degrees of freedom, should be > 0.\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            Drawn samples.\n\n        Notes\n        -----\n        The probability density function for the t distribution is\n\n        .. math:: P(x, df) = \\frac{\\Gamma(\\frac{df+1}{2})}{\\sqrt{\\pi df}\n                  \\Gamma(\\frac{df}{2})}\\Bigl( 1+\\frac{x^2}{df} \\Bigr)^{-(df+1)/2}\n\n        The t test is based on an assumption that the data come from a Normal\n        distribution. The t test provides a way to test whether the sample mean\n        (that is the mean calculated from the data) is a good estimate of the true\n        mean.\n\n        The derivation of the t-distribution was forst published in 1908 by William\n        Gisset while working for the Guinness Brewery in Dublin. Due to proprietary\n        issues, he had to publish under a pseudonym, and so he used the name\n        Student.\n\n        References\n        ----------\n        .. [1] Dalgaard, Peter, \"Introductory Statistics With R\",\n               Springer, 2002.\n        .. [2] Wikipedia, \"Student's t-distribution\"\n               http://en.wikipedia.org/wiki/Student's_t-distribution\n\n        Examples\n        --------\n        From Dalgaard page 83 [1]_, suppose the daily energy intake for 11\n        women in Kj is:\n\n        >>> intake = np.array([5260., 5470, 5640, 6180, 6390, 6515, 6805, 7515, \\\n        ...                    7515, 8230, 8770])\n\n        Doe""s their energy intake deviate systematically from the recommended\n        value of 7725 kJ?\n\n        We have 10 degrees of freedom, so is the sample mean within 95% of the\n        recommended value?\n\n        >>> s = np.random.standard_t(10, size=100000)\n        >>> np.mean(intake)\n        6753.636363636364\n        >>> intake.std(ddof=1)\n        1142.1232221373727\n\n        Calculate the t statistic, setting the ddof parameter to the unbiased\n        value so the divisor in the standard deviation will be degrees of\n        freedom, N-1.\n\n        >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake)))\n        >>> import matplotlib.pyplot as plt\n        >>> h = plt.hist(s, bins=100, normed=True)\n\n        For a one-sided t-test, how far out in the distribution does the t\n        statistic appear?\n\n        >>> >>> np.sum(s<t) / float(len(s))\n        0.0090699999999999999  #random\n\n        So the p-value is about 0.009, which says the null hypothesis has a\n        probability of about 99% of being true.\n\n        ";
-static char __pyx_k_246[] = "RandomState.vonmises (line 2345)";
-static char __pyx_k_247[] = "\n        vonmises(mu, kappa, size=None)\n\n        Draw samples from a von Mises distribution.\n\n        Samples are drawn from a von Mises distribution with specified mode\n        (mu) and dispersion (kappa), on the interval [-pi, pi].\n\n        The von Mises distribution (also known as the circular normal\n        distribution) is a continuous probability distribution on the unit\n        circle.  It may be thought of as the circular analogue of the normal\n        distribution.\n\n        Parameters\n        ----------\n        mu : float\n            Mode (\"center\") of the distribution.\n        kappa : float\n            Dispersion of the distribution, has to be >=0.\n        size : int or tuple of int\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : scalar or ndarray\n            The returned samples, which are in the interval [-pi, pi].\n\n        See Also\n        --------\n        scipy.stats.distributions.vonmises : probability density function,\n            distribution, or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the von Mises distribution is\n\n        .. math:: p(x) = \\frac{e^{\\kappa cos(x-\\mu)}}{2\\pi I_0(\\kappa)},\n\n        where :math:`\\mu` is the mode and :math:`\\kappa` the dispersion,\n        and :math:`I_0(\\kappa)` is the modified Bessel function of order 0.\n\n        The von Mises is named for Richard Edler von Mises, who was born in\n        Austria-Hungary, in what is now the Ukraine.  He fled to the United\n        States in 1939 and became a professor at Harvard.  He worked in\n        probability theory, aerodynamics, fluid mechanics, and philosophy of\n        science.\n\n        References\n        ----------\n        Abramowitz, M. and Stegun, I. A. (ed.), *Handbook of Mathematical\n        Functions*, New York: Dover, 1965.\n\n      ""  von Mises, R., *Mathematical Theory of Probability and Statistics*,\n        New York: Academic Press, 1964.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, kappa = 0.0, 4.0 # mean and dispersion\n        >>> s = np.random.vonmises(mu, kappa, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> import scipy.special as sps\n        >>> count, bins, ignored = plt.hist(s, 50, normed=True)\n        >>> x = np.arange(-np.pi, np.pi, 2*np.pi/50.)\n        >>> y = -np.exp(kappa*np.cos(x-mu))/(2*np.pi*sps.jn(0,kappa))\n        >>> plt.plot(x, y/max(y), linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_248[] = "RandomState.pareto (line 2439)";
-static char __pyx_k_249[] = "\n        pareto(a, size=None)\n\n        Draw samples from a Pareto II or Lomax distribution with specified shape.\n\n        The Lomax or Pareto II distribution is a shifted Pareto distribution. The\n        classical Pareto distribution can be obtained from the Lomax distribution\n        by adding the location parameter m, see below. The smallest value of the\n        Lomax distribution is zero while for the classical Pareto distribution it\n        is m, where the standard Pareto distribution has location m=1.\n        Lomax can also be considered as a simplified version of the Generalized\n        Pareto distribution (available in SciPy), with the scale set to one and\n        the location set to zero.\n\n        The Pareto distribution must be greater than zero, and is unbounded above.\n        It is also known as the \"80-20 rule\".  In this distribution, 80 percent of\n        the weights are in the lowest 20 percent of the range, while the other 20\n        percent fill the remaining 80 percent of the range.\n\n        Parameters\n        ----------\n        shape : float, > 0.\n            Shape of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        See Also\n        --------\n        scipy.stats.distributions.lomax.pdf : probability density function,\n            distribution or cumulative density function, etc.\n        scipy.stats.distributions.genpareto.pdf : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Pareto distribution is\n\n        .. math:: p(x) = \\frac{am^a}{x^{a+1}}\n\n        where :math:`a` is the shape and :math:`m` the location\n\n        The Pareto distribution, named after the Italian economist Vilfredo Pareto,\n        is a power law probability distribution useful in many real world probl""ems.\n        Outside the field of economics it is generally referred to as the Bradford\n        distribution. Pareto developed the distribution to describe the\n        distribution of wealth in an economy.  It has also found use in insurance,\n        web page access statistics, oil field sizes, and many other problems,\n        including the download frequency for projects in Sourceforge [1].  It is\n        one of the so-called \"fat-tailed\" distributions.\n\n\n        References\n        ----------\n        .. [1] Francis Hunt and Paul Johnson, On the Pareto Distribution of\n               Sourceforge projects.\n        .. [2] Pareto, V. (1896). Course of Political Economy. Lausanne.\n        .. [3] Reiss, R.D., Thomas, M.(2001), Statistical Analysis of Extreme\n               Values, Birkhauser Verlag, Basel, pp 23-30.\n        .. [4] Wikipedia, \"Pareto distribution\",\n               http://en.wikipedia.org/wiki/Pareto_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a, m = 3., 1. # shape and mode\n        >>> s = np.random.pareto(a, 1000) + m\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 100, normed=True, align='center')\n        >>> fit = a*m**a/bins**(a+1)\n        >>> plt.plot(bins, max(count)*fit/max(fit),linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_250[] = "RandomState.weibull (line 2535)";
-static char __pyx_k_251[] = "\n        weibull(a, size=None)\n\n        Weibull distribution.\n\n        Draw samples from a 1-parameter Weibull distribution with the given\n        shape parameter `a`.\n\n        .. math:: X = (-ln(U))^{1/a}\n\n        Here, U is drawn from the uniform distribution over (0,1].\n\n        The more common 2-parameter Weibull, including a scale parameter\n        :math:`\\lambda` is just :math:`X = \\lambda(-ln(U))^{1/a}`.\n\n        Parameters\n        ----------\n        a : float\n            Shape of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        See Also\n        --------\n        scipy.stats.distributions.weibull_max\n        scipy.stats.distributions.weibull_min\n        scipy.stats.distributions.genextreme\n        gumbel\n\n        Notes\n        -----\n        The Weibull (or Type III asymptotic extreme value distribution for smallest\n        values, SEV Type III, or Rosin-Rammler distribution) is one of a class of\n        Generalized Extreme Value (GEV) distributions used in modeling extreme\n        value problems.  This class includes the Gumbel and Frechet distributions.\n\n        The probability density for the Weibull distribution is\n\n        .. math:: p(x) = \\frac{a}\n                         {\\lambda}(\\frac{x}{\\lambda})^{a-1}e^{-(x/\\lambda)^a},\n\n        where :math:`a` is the shape and :math:`\\lambda` the scale.\n\n        The function has its peak (the mode) at\n        :math:`\\lambda(\\frac{a-1}{a})^{1/a}`.\n\n        When ``a = 1``, the Weibull distribution reduces to the exponential\n        distribution.\n\n        References\n        ----------\n        .. [1] Waloddi Weibull, Professor, Royal Technical University, Stockholm,\n               1939 \"A Statistical Theory Of The Strength Of Materials\",\n               Ingeniorsvetenskapsakademiens Handlingar Nr 151, 1939,\n               General""stabens Litografiska Anstalts Forlag, Stockholm.\n        .. [2] Waloddi Weibull, 1951 \"A Statistical Distribution Function of Wide\n               Applicability\",  Journal Of Applied Mechanics ASME Paper.\n        .. [3] Wikipedia, \"Weibull distribution\",\n               http://en.wikipedia.org/wiki/Weibull_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = 5. # shape\n        >>> s = np.random.weibull(a, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> x = np.arange(1,100.)/50.\n        >>> def weib(x,n,a):\n        ...     return (a / n) * (x / n)**(a - 1) * np.exp(-(x / n)**a)\n\n        >>> count, bins, ignored = plt.hist(np.random.weibull(5.,1000))\n        >>> x = np.arange(1,100.)/50.\n        >>> scale = count.max()/weib(x, 1., 5.).max()\n        >>> plt.plot(x, weib(x, 1., 5.)*scale)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_252[] = "RandomState.power (line 2635)";
-static char __pyx_k_253[] = "\n        power(a, size=None)\n\n        Draws samples in [0, 1] from a power distribution with positive\n        exponent a - 1.\n\n        Also known as the power function distribution.\n\n        Parameters\n        ----------\n        a : float\n            parameter, > 0\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n                    ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n            The returned samples lie in [0, 1].\n\n        Raises\n        ------\n        ValueError\n            If a<1.\n\n        Notes\n        -----\n        The probability density function is\n\n        .. math:: P(x; a) = ax^{a-1}, 0 \\le x \\le 1, a>0.\n\n        The power function distribution is just the inverse of the Pareto\n        distribution. It may also be seen as a special case of the Beta\n        distribution.\n\n        It is used, for example, in modeling the over-reporting of insurance\n        claims.\n\n        References\n        ----------\n        .. [1] Christian Kleiber, Samuel Kotz, \"Statistical size distributions\n               in economics and actuarial sciences\", Wiley, 2003.\n        .. [2] Heckert, N. A. and Filliben, James J. (2003). NIST Handbook 148:\n               Dataplot Reference Manual, Volume 2: Let Subcommands and Library\n               Functions\", National Institute of Standards and Technology Handbook\n               Series, June 2003.\n               http://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = 5. # shape\n        >>> samples = 1000\n        >>> s = np.random.power(a, samples)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, bins=""30)\n        >>> x = np.linspace(0, 1, 100)\n        >>> y = a*x**(a-1.)\n        >>> normed_y = samples*np.diff(bins)[0]*y\n        >>> plt.plot(x, normed_y)\n        >>> plt.show()\n\n        Compare the power function distribution to the inverse of the Pareto.\n\n        >>> from scipy import stats\n        >>> rvs = np.random.power(5, 1000000)\n        >>> rvsp = np.random.pareto(5, 1000000)\n        >>> xx = np.linspace(0,1,100)\n        >>> powpdf = stats.powerlaw.pdf(xx,5)\n\n        >>> plt.figure()\n        >>> plt.hist(rvs, bins=50, normed=True)\n        >>> plt.plot(xx,powpdf,'r-')\n        >>> plt.title('np.random.power(5)')\n\n        >>> plt.figure()\n        >>> plt.hist(1./(1.+rvsp), bins=50, normed=True)\n        >>> plt.plot(xx,powpdf,'r-')\n        >>> plt.title('inverse of 1 + np.random.pareto(5)')\n\n        >>> plt.figure()\n        >>> plt.hist(1./(1.+rvsp), bins=50, normed=True)\n        >>> plt.plot(xx,powpdf,'r-')\n        >>> plt.title('inverse of stats.pareto(5)')\n\n        ";
-static char __pyx_k_254[] = "RandomState.laplace (line 2744)";
-static char __pyx_k_255[] = "\n        laplace(loc=0.0, scale=1.0, size=None)\n\n        Draw samples from the Laplace or double exponential distribution with\n        specified location (or mean) and scale (decay).\n\n        The Laplace distribution is similar to the Gaussian/normal distribution,\n        but is sharper at the peak and has fatter tails. It represents the\n        difference between two independent, identically distributed exponential\n        random variables.\n\n        Parameters\n        ----------\n        loc : float\n            The position, :math:`\\mu`, of the distribution peak.\n        scale : float\n            :math:`\\lambda`, the exponential decay.\n\n        Notes\n        -----\n        It has the probability density function\n\n        .. math:: f(x; \\mu, \\lambda) = \\frac{1}{2\\lambda}\n                                       \\exp\\left(-\\frac{|x - \\mu|}{\\lambda}\\right).\n\n        The first law of Laplace, from 1774, states that the frequency of an error\n        can be expressed as an exponential function of the absolute magnitude of\n        the error, which leads to the Laplace distribution. For many problems in\n        Economics and Health sciences, this distribution seems to model the data\n        better than the standard Gaussian distribution\n\n\n        References\n        ----------\n        .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). Handbook of Mathematical\n               Functions with Formulas, Graphs, and Mathematical Tables, 9th\n               printing.  New York: Dover, 1972.\n\n        .. [2] The Laplace distribution and generalizations\n               By Samuel Kotz, Tomasz J. Kozubowski, Krzysztof Podgorski,\n               Birkhauser, 2001.\n\n        .. [3] Weisstein, Eric W. \"Laplace Distribution.\"\n               From MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/LaplaceDistribution.html\n\n        .. [4] Wikipedia, \"Laplace distribution\",\n               http://en.wikipedia.org/wik""i/Laplace_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution\n\n        >>> loc, scale = 0., 1.\n        >>> s = np.random.laplace(loc, scale, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 30, normed=True)\n        >>> x = np.arange(-8., 8., .01)\n        >>> pdf = np.exp(-abs(x-loc)/scale)/(2.*scale)\n        >>> plt.plot(x, pdf)\n\n        Plot Gaussian for comparison:\n\n        >>> g = (1/(scale * np.sqrt(2 * np.pi)) * \n        ...      np.exp(-(x - loc)**2 / (2 * scale**2)))\n        >>> plt.plot(x,g)\n\n        ";
-static char __pyx_k_256[] = "RandomState.gumbel (line 2834)";
-static char __pyx_k_257[] = "\n        gumbel(loc=0.0, scale=1.0, size=None)\n\n        Gumbel distribution.\n\n        Draw samples from a Gumbel distribution with specified location and scale.\n        For more information on the Gumbel distribution, see Notes and References\n        below.\n\n        Parameters\n        ----------\n        loc : float\n            The location of the mode of the distribution.\n        scale : float\n            The scale parameter of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        out : ndarray\n            The samples\n\n        See Also\n        --------\n        scipy.stats.gumbel_l\n        scipy.stats.gumbel_r\n        scipy.stats.genextreme\n            probability density function, distribution, or cumulative density\n            function, etc. for each of the above\n        weibull\n\n        Notes\n        -----\n        The Gumbel (or Smallest Extreme Value (SEV) or the Smallest Extreme Value\n        Type I) distribution is one of a class of Generalized Extreme Value (GEV)\n        distributions used in modeling extreme value problems.  The Gumbel is a\n        special case of the Extreme Value Type I distribution for maximums from\n        distributions with \"exponential-like\" tails.\n\n        The probability density for the Gumbel distribution is\n\n        .. math:: p(x) = \\frac{e^{-(x - \\mu)/ \\beta}}{\\beta} e^{ -e^{-(x - \\mu)/\n                  \\beta}},\n\n        where :math:`\\mu` is the mode, a location parameter, and :math:`\\beta` is\n        the scale parameter.\n\n        The Gumbel (named for German mathematician Emil Julius Gumbel) was used\n        very early in the hydrology literature, for modeling the occurrence of\n        flood events. It is also used for modeling maximum wind speed and rainfall\n        rates.  It is a \"fat-tailed\" distribution - the ""probability of an event in\n        the tail of the distribution is larger than if one used a Gaussian, hence\n        the surprisingly frequent occurrence of 100-year floods. Floods were\n        initially modeled as a Gaussian process, which underestimated the frequency\n        of extreme events.\n\n\n        It is one of a class of extreme value distributions, the Generalized\n        Extreme Value (GEV) distributions, which also includes the Weibull and\n        Frechet.\n\n        The function has a mean of :math:`\\mu + 0.57721\\beta` and a variance of\n        :math:`\\frac{\\pi^2}{6}\\beta^2`.\n\n        References\n        ----------\n        Gumbel, E. J., *Statistics of Extremes*, New York: Columbia University\n        Press, 1958.\n\n        Reiss, R.-D. and Thomas, M., *Statistical Analysis of Extreme Values from\n        Insurance, Finance, Hydrology and Other Fields*, Basel: Birkhauser Verlag,\n        2001.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, beta = 0, 0.1 # location and scale\n        >>> s = np.random.gumbel(mu, beta, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 30, normed=True)\n        >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta)\n        ...          * np.exp( -np.exp( -(bins - mu) /beta) ),\n        ...          linewidth=2, color='r')\n        >>> plt.show()\n\n        Show how an extreme value distribution can arise from a Gaussian process\n        and compare to a Gaussian:\n\n        >>> means = []\n        >>> maxima = []\n        >>> for i in range(0,1000) :\n        ...    a = np.random.normal(mu, beta, 1000)\n        ...    means.append(a.mean())\n        ...    maxima.append(a.max())\n        >>> count, bins, ignored = plt.hist(maxima, 30, normed=True)\n        >>> beta = np.std(maxima)*np.pi/np.sqrt(6)""\n        >>> mu = np.mean(maxima) - 0.57721*beta\n        >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta)\n        ...          * np.exp(-np.exp(-(bins - mu)/beta)),\n        ...          linewidth=2, color='r')\n        >>> plt.plot(bins, 1/(beta * np.sqrt(2 * np.pi))\n        ...          * np.exp(-(bins - mu)**2 / (2 * beta**2)),\n        ...          linewidth=2, color='g')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_258[] = "RandomState.logistic (line 2965)";
-static char __pyx_k_259[] = "\n        logistic(loc=0.0, scale=1.0, size=None)\n\n        Draw samples from a Logistic distribution.\n\n        Samples are drawn from a Logistic distribution with specified\n        parameters, loc (location or mean, also median), and scale (>0).\n\n        Parameters\n        ----------\n        loc : float\n\n        scale : float > 0.\n\n        size : {tuple, int}\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n                  where the values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.logistic : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Logistic distribution is\n\n        .. math:: P(x) = P(x) = \\frac{e^{-(x-\\mu)/s}}{s(1+e^{-(x-\\mu)/s})^2},\n\n        where :math:`\\mu` = location and :math:`s` = scale.\n\n        The Logistic distribution is used in Extreme Value problems where it\n        can act as a mixture of Gumbel distributions, in Epidemiology, and by\n        the World Chess Federation (FIDE) where it is used in the Elo ranking\n        system, assuming the performance of each player is a logistically\n        distributed random variable.\n\n        References\n        ----------\n        .. [1] Reiss, R.-D. and Thomas M. (2001), Statistical Analysis of Extreme\n               Values, from Insurance, Finance, Hydrology and Other Fields,\n               Birkhauser Verlag, Basel, pp 132-133.\n        .. [2] Weisstein, Eric W. \"Logistic Distribution.\" From\n               MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/LogisticDistribution.html\n        .. [3] Wikipedia, \"Logistic-distribution\",\n               http://en.wikipedia.org/wiki/Logistic-distribution\n\n        Examples\n   ""     --------\n        Draw samples from the distribution:\n\n        >>> loc, scale = 10, 1\n        >>> s = np.random.logistic(loc, scale, 10000)\n        >>> count, bins, ignored = plt.hist(s, bins=50)\n\n        #   plot against distribution\n\n        >>> def logist(x, loc, scale):\n        ...     return exp((loc-x)/scale)/(scale*(1+exp((loc-x)/scale))**2)\n        >>> plt.plot(bins, logist(bins, loc, scale)*count.max()/\\\n        ... logist(bins, loc, scale).max())\n        >>> plt.show()\n\n        ";
-static char __pyx_k_260[] = "RandomState.lognormal (line 3053)";
-static char __pyx_k_261[] = "\n        lognormal(mean=0.0, sigma=1.0, size=None)\n\n        Return samples drawn from a log-normal distribution.\n\n        Draw samples from a log-normal distribution with specified mean,\n        standard deviation, and array shape.  Note that the mean and standard\n        deviation are not the values for the distribution itself, but of the\n        underlying normal distribution it is derived from.\n\n        Parameters\n        ----------\n        mean : float\n            Mean value of the underlying normal distribution\n        sigma : float, > 0.\n            Standard deviation of the underlying normal distribution\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : ndarray or float\n            The desired samples. An array of the same shape as `size` if given,\n            if `size` is None a float is returned.\n\n        See Also\n        --------\n        scipy.stats.lognorm : probability density function, distribution,\n            cumulative density function, etc.\n\n        Notes\n        -----\n        A variable `x` has a log-normal distribution if `log(x)` is normally\n        distributed.  The probability density function for the log-normal\n        distribution is:\n\n        .. math:: p(x) = \\frac{1}{\\sigma x \\sqrt{2\\pi}}\n                         e^{(-\\frac{(ln(x)-\\mu)^2}{2\\sigma^2})}\n\n        where :math:`\\mu` is the mean and :math:`\\sigma` is the standard\n        deviation of the normally distributed logarithm of the variable.\n        A log-normal distribution results if a random variable is the *product*\n        of a large number of independent, identically-distributed variables in\n        the same way that a normal distribution results if the variable is the\n        *sum* of a large number of independent, identically-distributed\n        variables.\n\n        Reference""s\n        ----------\n        Limpert, E., Stahel, W. A., and Abbt, M., \"Log-normal Distributions\n        across the Sciences: Keys and Clues,\" *BioScience*, Vol. 51, No. 5,\n        May, 2001.  http://stat.ethz.ch/~stahel/lognormal/bioscience.pdf\n\n        Reiss, R.D. and Thomas, M., *Statistical Analysis of Extreme Values*,\n        Basel: Birkhauser Verlag, 2001, pp. 31-32.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, sigma = 3., 1. # mean and standard deviation\n        >>> s = np.random.lognormal(mu, sigma, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 100, normed=True, align='mid')\n\n        >>> x = np.linspace(min(bins), max(bins), 10000)\n        >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2))\n        ...        / (x * sigma * np.sqrt(2 * np.pi)))\n\n        >>> plt.plot(x, pdf, linewidth=2, color='r')\n        >>> plt.axis('tight')\n        >>> plt.show()\n\n        Demonstrate that taking the products of random samples from a uniform\n        distribution can be fit well by a log-normal probability density function.\n\n        >>> # Generate a thousand samples: each is the product of 100 random\n        >>> # values, drawn from a normal distribution.\n        >>> b = []\n        >>> for i in range(1000):\n        ...    a = 10. + np.random.random(100)\n        ...    b.append(np.product(a))\n\n        >>> b = np.array(b) / np.min(b) # scale values to be positive\n        >>> count, bins, ignored = plt.hist(b, 100, normed=True, align='center')\n        >>> sigma = np.std(np.log(b))\n        >>> mu = np.mean(np.log(b))\n\n        >>> x = np.linspace(min(bins), max(bins), 10000)\n        >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2))\n        ...        / (x * sigma * np.sqrt(2 * np.pi)))\n\n        >>> plt.plot(x, pdf, co""lor='r', linewidth=2)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_262[] = "RandomState.rayleigh (line 3174)";
-static char __pyx_k_263[] = "\n        rayleigh(scale=1.0, size=None)\n\n        Draw samples from a Rayleigh distribution.\n\n        The :math:`\\chi` and Weibull distributions are generalizations of the\n        Rayleigh.\n\n        Parameters\n        ----------\n        scale : scalar\n            Scale, also equals the mode. Should be >= 0.\n        size : int or tuple of ints, optional\n            Shape of the output. Default is None, in which case a single\n            value is returned.\n\n        Notes\n        -----\n        The probability density function for the Rayleigh distribution is\n\n        .. math:: P(x;scale) = \\frac{x}{scale^2}e^{\\frac{-x^2}{2 \\cdotp scale^2}}\n\n        The Rayleigh distribution arises if the wind speed and wind direction are\n        both gaussian variables, then the vector wind velocity forms a Rayleigh\n        distribution. The Rayleigh distribution is used to model the expected\n        output from wind turbines.\n\n        References\n        ----------\n        .. [1] Brighton Webs Ltd., Rayleigh Distribution,\n              http://www.brighton-webs.co.uk/distributions/rayleigh.asp\n        .. [2] Wikipedia, \"Rayleigh distribution\"\n              http://en.wikipedia.org/wiki/Rayleigh_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram\n\n        >>> values = hist(np.random.rayleigh(3, 100000), bins=200, normed=True)\n\n        Wave heights tend to follow a Rayleigh distribution. If the mean wave\n        height is 1 meter, what fraction of waves are likely to be larger than 3\n        meters?\n\n        >>> meanvalue = 1\n        >>> modevalue = np.sqrt(2 / np.pi) * meanvalue\n        >>> s = np.random.rayleigh(modevalue, 1000000)\n\n        The percentage of waves larger than 3 meters is:\n\n        >>> 100.*sum(s>3)/1000000.\n        0.087300000000000003\n\n        ";
-static char __pyx_k_264[] = "RandomState.wald (line 3246)";
-static char __pyx_k_265[] = "\n        wald(mean, scale, size=None)\n\n        Draw samples from a Wald, or Inverse Gaussian, distribution.\n\n        As the scale approaches infinity, the distribution becomes more like a\n        Gaussian.\n\n        Some references claim that the Wald is an Inverse Gaussian with mean=1, but\n        this is by no means universal.\n\n        The Inverse Gaussian distribution was first studied in relationship to\n        Brownian motion. In 1956 M.C.K. Tweedie used the name Inverse Gaussian\n        because there is an inverse relationship between the time to cover a unit\n        distance and distance covered in unit time.\n\n        Parameters\n        ----------\n        mean : scalar\n            Distribution mean, should be > 0.\n        scale : scalar\n            Scale parameter, should be >= 0.\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            Drawn sample, all greater than zero.\n\n        Notes\n        -----\n        The probability density function for the Wald distribution is\n\n        .. math:: P(x;mean,scale) = \\sqrt{\\frac{scale}{2\\pi x^3}}e^\n                                    \\frac{-scale(x-mean)^2}{2\\cdotp mean^2x}\n\n        As noted above the Inverse Gaussian distribution first arise from attempts\n        to model Brownian Motion. It is also a competitor to the Weibull for use in\n        reliability modeling and modeling stock returns and interest rate\n        processes.\n\n        References\n        ----------\n        .. [1] Brighton Webs Ltd., Wald Distribution,\n              http://www.brighton-webs.co.uk/distributions/wald.asp\n        .. [2] Chhikara, Raj S., and Folks, J. Leroy, \"The Inverse Gaussian\n              Distribution: Theory : Methodology, and Applications\", CRC Press,\n              1988.\n        .. [3] Wikipedia, \"Wald distribu""tion\"\n              http://en.wikipedia.org/wiki/Wald_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram:\n\n        >>> import matplotlib.pyplot as plt\n        >>> h = plt.hist(np.random.wald(3, 2, 100000), bins=200, normed=True)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_266[] = "RandomState.triangular (line 3332)";
-static char __pyx_k_267[] = "\n        triangular(left, mode, right, size=None)\n\n        Draw samples from the triangular distribution.\n\n        The triangular distribution is a continuous probability distribution with\n        lower limit left, peak at mode, and upper limit right. Unlike the other\n        distributions, these parameters directly define the shape of the pdf.\n\n        Parameters\n        ----------\n        left : scalar\n            Lower limit.\n        mode : scalar\n            The value where the peak of the distribution occurs.\n            The value should fulfill the condition ``left <= mode <= right``.\n        right : scalar\n            Upper limit, should be larger than `left`.\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The returned samples all lie in the interval [left, right].\n\n        Notes\n        -----\n        The probability density function for the Triangular distribution is\n\n        .. math:: P(x;l, m, r) = \\begin{cases}\n                  \\frac{2(x-l)}{(r-l)(m-l)}& \\text{for $l \\leq x \\leq m$},\\\\\n                  \\frac{2(m-x)}{(r-l)(r-m)}& \\text{for $m \\leq x \\leq r$},\\\\\n                  0& \\text{otherwise}.\n                  \\end{cases}\n\n        The triangular distribution is often used in ill-defined problems where the\n        underlying distribution is not known, but some knowledge of the limits and\n        mode exists. Often it is used in simulations.\n\n        References\n        ----------\n        .. [1] Wikipedia, \"Triangular distribution\"\n              http://en.wikipedia.org/wiki/Triangular_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram:\n\n        >>> import matplotlib.pyplot as plt\n        >>> h = plt.hist(np.random.triangular(-3, 0, 8, 100000), bins=""200,\n        ...              normed=True)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_268[] = "RandomState.binomial (line 3420)";
-static char __pyx_k_269[] = "\n        binomial(n, p, size=None)\n\n        Draw samples from a binomial distribution.\n\n        Samples are drawn from a Binomial distribution with specified\n        parameters, n trials and p probability of success where\n        n an integer >= 0 and p is in the interval [0,1]. (n may be\n        input as a float, but it is truncated to an integer in use)\n\n        Parameters\n        ----------\n        n : float (but truncated to an integer)\n                parameter, >= 0.\n        p : float\n                parameter, >= 0 and <=1.\n        size : {tuple, int}\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n                  where the values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.binom : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Binomial distribution is\n\n        .. math:: P(N) = \\binom{n}{N}p^N(1-p)^{n-N},\n\n        where :math:`n` is the number of trials, :math:`p` is the probability\n        of success, and :math:`N` is the number of successes.\n\n        When estimating the standard error of a proportion in a population by\n        using a random sample, the normal distribution works well unless the\n        product p*n <=5, where p = population proportion estimate, and n =\n        number of samples, in which case the binomial distribution is used\n        instead. For example, a sample of 15 people shows 4 who are left\n        handed, and 11 who are right handed. Then p = 4/15 = 27%. 0.27*15 = 4,\n        so the binomial distribution should be used in this case.\n\n        References\n        ----------\n        .. [1] Dalgaard, Peter, \"Introductory Statistics with R\",\n               Springer-Verlag, 2002.""\n        .. [2] Glantz, Stanton A. \"Primer of Biostatistics.\", McGraw-Hill,\n               Fifth Edition, 2002.\n        .. [3] Lentner, Marvin, \"Elementary Applied Statistics\", Bogden\n               and Quigley, 1972.\n        .. [4] Weisstein, Eric W. \"Binomial Distribution.\" From MathWorld--A\n               Wolfram Web Resource.\n               http://mathworld.wolfram.com/BinomialDistribution.html\n        .. [5] Wikipedia, \"Binomial-distribution\",\n               http://en.wikipedia.org/wiki/Binomial_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> n, p = 10, .5 # number of trials, probability of each trial\n        >>> s = np.random.binomial(n, p, 1000)\n        # result of flipping a coin 10 times, tested 1000 times.\n\n        A real world example. A company drills 9 wild-cat oil exploration\n        wells, each with an estimated probability of success of 0.1. All nine\n        wells fail. What is the probability of that happening?\n\n        Let's do 20,000 trials of the model, and count the number that\n        generate zero positive results.\n\n        >>> sum(np.random.binomial(9,0.1,20000)==0)/20000.\n        answer = 0.38885, or 38%.\n\n        ";
-static char __pyx_k_270[] = "RandomState.negative_binomial (line 3528)";
-static char __pyx_k_271[] = "\n        negative_binomial(n, p, size=None)\n\n        Draw samples from a negative_binomial distribution.\n\n        Samples are drawn from a negative_Binomial distribution with specified\n        parameters, `n` trials and `p` probability of success where `n` is an\n        integer > 0 and `p` is in the interval [0, 1].\n\n        Parameters\n        ----------\n        n : int\n            Parameter, > 0.\n        p : float\n            Parameter, >= 0 and <=1.\n        size : int or tuple of ints\n            Output shape. If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : int or ndarray of ints\n            Drawn samples.\n\n        Notes\n        -----\n        The probability density for the Negative Binomial distribution is\n\n        .. math:: P(N;n,p) = \\binom{N+n-1}{n-1}p^{n}(1-p)^{N},\n\n        where :math:`n-1` is the number of successes, :math:`p` is the probability\n        of success, and :math:`N+n-1` is the number of trials.\n\n        The negative binomial distribution gives the probability of n-1 successes\n        and N failures in N+n-1 trials, and success on the (N+n)th trial.\n\n        If one throws a die repeatedly until the third time a \"1\" appears, then the\n        probability distribution of the number of non-\"1\"s that appear before the\n        third \"1\" is a negative binomial distribution.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Negative Binomial Distribution.\" From\n               MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/NegativeBinomialDistribution.html\n        .. [2] Wikipedia, \"Negative binomial distribution\",\n               http://en.wikipedia.org/wiki/Negative_binomial_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        A real world example. A company drills wild-cat oil exploration well""s, each\n        with an estimated probability of success of 0.1.  What is the probability\n        of having one success for each successive well, that is what is the\n        probability of a single success after drilling 5 wells, after 6 wells,\n        etc.?\n\n        >>> s = np.random.negative_binomial(1, 0.1, 100000)\n        >>> for i in range(1, 11):\n        ...    probability = sum(s<i) / 100000.\n        ...    print i, \"wells drilled, probability of one success =\", probability\n\n        ";
-static char __pyx_k_272[] = "RandomState.poisson (line 3623)";
-static char __pyx_k_273[] = "\n        poisson(lam=1.0, size=None)\n\n        Draw samples from a Poisson distribution.\n\n        The Poisson distribution is the limit of the Binomial\n        distribution for large N.\n\n        Parameters\n        ----------\n        lam : float\n            Expectation of interval, should be >= 0.\n        size : int or tuple of ints, optional\n            Output shape. If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Notes\n        -----\n        The Poisson distribution\n\n        .. math:: f(k; \\lambda)=\\frac{\\lambda^k e^{-\\lambda}}{k!}\n\n        For events with an expected separation :math:`\\lambda` the Poisson\n        distribution :math:`f(k; \\lambda)` describes the probability of\n        :math:`k` events occurring within the observed interval :math:`\\lambda`.\n\n        Because the output is limited to the range of the C long type, a\n        ValueError is raised when `lam` is within 10 sigma of the maximum\n        representable value.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Poisson Distribution.\" From MathWorld--A Wolfram\n               Web Resource. http://mathworld.wolfram.com/PoissonDistribution.html\n        .. [2] Wikipedia, \"Poisson distribution\",\n           http://en.wikipedia.org/wiki/Poisson_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> import numpy as np\n        >>> s = np.random.poisson(5, 10000)\n\n        Display histogram of the sample:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 14, normed=True)\n        >>> plt.show()\n\n        ";
-static char __pyx_k_274[] = "RandomState.zipf (line 3694)";
-static char __pyx_k_275[] = "\n        zipf(a, size=None)\n\n        Draw samples from a Zipf distribution.\n\n        Samples are drawn from a Zipf distribution with specified parameter\n        `a` > 1.\n\n        The Zipf distribution (also known as the zeta distribution) is a\n        continuous probability distribution that satisfies Zipf's law: the\n        frequency of an item is inversely proportional to its rank in a\n        frequency table.\n\n        Parameters\n        ----------\n        a : float > 1\n            Distribution parameter.\n        size : int or tuple of int, optional\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn; a single integer is equivalent in\n            its result to providing a mono-tuple, i.e., a 1-D array of length\n            *size* is returned.  The default is None, in which case a single\n            scalar is returned.\n\n        Returns\n        -------\n        samples : scalar or ndarray\n            The returned samples are greater than or equal to one.\n\n        See Also\n        --------\n        scipy.stats.distributions.zipf : probability density function,\n            distribution, or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Zipf distribution is\n\n        .. math:: p(x) = \\frac{x^{-a}}{\\zeta(a)},\n\n        where :math:`\\zeta` is the Riemann Zeta function.\n\n        It is named for the American linguist George Kingsley Zipf, who noted\n        that the frequency of any word in a sample of a language is inversely\n        proportional to its rank in the frequency table.\n\n        References\n        ----------\n        Zipf, G. K., *Selected Studies of the Principle of Relative Frequency\n        in Language*, Cambridge, MA: Harvard Univ. Press, 1932.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = 2. # parameter\n        >>> s = np.random.zipf""(a, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> import scipy.special as sps\n        Truncate s values at 50 so plot is interesting\n        >>> count, bins, ignored = plt.hist(s[s<50], 50, normed=True)\n        >>> x = np.arange(1., 50.)\n        >>> y = x**(-a)/sps.zetac(a)\n        >>> plt.plot(x, y/max(y), linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_276[] = "RandomState.geometric (line 3782)";
-static char __pyx_k_277[] = "\n        geometric(p, size=None)\n\n        Draw samples from the geometric distribution.\n\n        Bernoulli trials are experiments with one of two outcomes:\n        success or failure (an example of such an experiment is flipping\n        a coin).  The geometric distribution models the number of trials\n        that must be run in order to achieve success.  It is therefore\n        supported on the positive integers, ``k = 1, 2, ...``.\n\n        The probability mass function of the geometric distribution is\n\n        .. math:: f(k) = (1 - p)^{k - 1} p\n\n        where `p` is the probability of success of an individual trial.\n\n        Parameters\n        ----------\n        p : float\n            The probability of success of an individual trial.\n        size : tuple of ints\n            Number of values to draw from the distribution.  The output\n            is shaped according to `size`.\n\n        Returns\n        -------\n        out : ndarray\n            Samples from the geometric distribution, shaped according to\n            `size`.\n\n        Examples\n        --------\n        Draw ten thousand values from the geometric distribution,\n        with the probability of an individual success equal to 0.35:\n\n        >>> z = np.random.geometric(p=0.35, size=10000)\n\n        How many trials succeeded after a single run?\n\n        >>> (z == 1).sum() / 10000.\n        0.34889999999999999 #random\n\n        ";
-static char __pyx_k_278[] = "RandomState.hypergeometric (line 3848)";
-static char __pyx_k_279[] = "\n        hypergeometric(ngood, nbad, nsample, size=None)\n\n        Draw samples from a Hypergeometric distribution.\n\n        Samples are drawn from a Hypergeometric distribution with specified\n        parameters, ngood (ways to make a good selection), nbad (ways to make\n        a bad selection), and nsample = number of items sampled, which is less\n        than or equal to the sum ngood + nbad.\n\n        Parameters\n        ----------\n        ngood : int or array_like\n            Number of ways to make a good selection.  Must be nonnegative.\n        nbad : int or array_like\n            Number of ways to make a bad selection.  Must be nonnegative.\n        nsample : int or array_like\n            Number of items sampled.  Must be at least 1 and at most\n            ``ngood + nbad``.\n        size : int or tuple of int\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.hypergeom : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Hypergeometric distribution is\n\n        .. math:: P(x) = \\frac{\\binom{m}{n}\\binom{N-m}{n-x}}{\\binom{N}{n}},\n\n        where :math:`0 \\le x \\le m` and :math:`n+m-N \\le x \\le n`\n\n        for P(x) the probability of x successes, n = ngood, m = nbad, and\n        N = number of samples.\n\n        Consider an urn with black and white marbles in it, ngood of them\n        black and nbad are white. If you draw nsample balls without\n        replacement, then the Hypergeometric distribution describes the\n        distribution of black balls in the drawn sample.\n\n        Note that this distribution is very similar to the Binomial\n        distrib""ution, except that in this case, samples are drawn without\n        replacement, whereas in the Binomial case samples are drawn with\n        replacement (or the sample space is infinite). As the sample space\n        becomes large, this distribution approaches the Binomial.\n\n        References\n        ----------\n        .. [1] Lentner, Marvin, \"Elementary Applied Statistics\", Bogden\n               and Quigley, 1972.\n        .. [2] Weisstein, Eric W. \"Hypergeometric Distribution.\" From\n               MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/HypergeometricDistribution.html\n        .. [3] Wikipedia, \"Hypergeometric-distribution\",\n               http://en.wikipedia.org/wiki/Hypergeometric-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> ngood, nbad, nsamp = 100, 2, 10\n        # number of good, number of bad, and number of samples\n        >>> s = np.random.hypergeometric(ngood, nbad, nsamp, 1000)\n        >>> hist(s)\n        #   note that it is very unlikely to grab both bad items\n\n        Suppose you have an urn with 15 white and 15 black marbles.\n        If you pull 15 marbles at random, how likely is it that\n        12 or more of them are one color?\n\n        >>> s = np.random.hypergeometric(15, 15, 15, 100000)\n        >>> sum(s>=12)/100000. + sum(s<=3)/100000.\n        #   answer = 0.003 ... pretty unlikely!\n\n        ";
-static char __pyx_k_280[] = "RandomState.logseries (line 3967)";
-static char __pyx_k_281[] = "\n        logseries(p, size=None)\n\n        Draw samples from a Logarithmic Series distribution.\n\n        Samples are drawn from a Log Series distribution with specified\n        parameter, p (probability, 0 < p < 1).\n\n        Parameters\n        ----------\n        loc : float\n\n        scale : float > 0.\n\n        size : {tuple, int}\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n                  where the values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.logser : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Log Series distribution is\n\n        .. math:: P(k) = \\frac{-p^k}{k \\ln(1-p)},\n\n        where p = probability.\n\n        The Log Series distribution is frequently used to represent species\n        richness and occurrence, first proposed by Fisher, Corbet, and\n        Williams in 1943 [2].  It may also be used to model the numbers of\n        occupants seen in cars [3].\n\n        References\n        ----------\n        .. [1] Buzas, Martin A.; Culver, Stephen J.,  Understanding regional\n               species diversity through the log series distribution of\n               occurrences: BIODIVERSITY RESEARCH Diversity & Distributions,\n               Volume 5, Number 5, September 1999 , pp. 187-195(9).\n        .. [2] Fisher, R.A,, A.S. Corbet, and C.B. Williams. 1943. The\n               relation between the number of species and the number of\n               individuals in a random sample of an animal population.\n               Journal of Animal Ecology, 12:42-58.\n        .. [3] D. J. Hand, F. Daly, D. Lunn, E. Ostrowski, A Handbook of Small\n               Data Sets, CRC Press, 1994.\n        .. [4] Wikipedia, \"Log""arithmic-distribution\",\n               http://en.wikipedia.org/wiki/Logarithmic-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = .6\n        >>> s = np.random.logseries(a, 10000)\n        >>> count, bins, ignored = plt.hist(s)\n\n        #   plot against distribution\n\n        >>> def logseries(k, p):\n        ...     return -p**k/(k*log(1-p))\n        >>> plt.plot(bins, logseries(bins, a)*count.max()/\n                     logseries(bins, a).max(), 'r')\n        >>> plt.show()\n\n        ";
-static char __pyx_k_282[] = "RandomState.multivariate_normal (line 4062)";
-static char __pyx_k_283[] = "\n        multivariate_normal(mean, cov[, size])\n\n        Draw random samples from a multivariate normal distribution.\n\n        The multivariate normal, multinormal or Gaussian distribution is a\n        generalization of the one-dimensional normal distribution to higher\n        dimensions.  Such a distribution is specified by its mean and\n        covariance matrix.  These parameters are analogous to the mean\n        (average or \"center\") and variance (standard deviation, or \"width,\"\n        squared) of the one-dimensional normal distribution.\n\n        Parameters\n        ----------\n        mean : 1-D array_like, of length N\n            Mean of the N-dimensional distribution.\n        cov : 2-D array_like, of shape (N, N)\n            Covariance matrix of the distribution.  Must be symmetric and\n            positive-semidefinite for \"physically meaningful\" results.\n        size : int or tuple of ints, optional\n            Given a shape of, for example, ``(m,n,k)``, ``m*n*k`` samples are\n            generated, and packed in an `m`-by-`n`-by-`k` arrangement.  Because\n            each sample is `N`-dimensional, the output shape is ``(m,n,k,N)``.\n            If no shape is specified, a single (`N`-D) sample is returned.\n\n        Returns\n        -------\n        out : ndarray\n            The drawn samples, of shape *size*, if that was provided.  If not,\n            the shape is ``(N,)``.\n\n            In other words, each entry ``out[i,j,...,:]`` is an N-dimensional\n            value drawn from the distribution.\n\n        Notes\n        -----\n        The mean is a coordinate in N-dimensional space, which represents the\n        location where samples are most likely to be generated.  This is\n        analogous to the peak of the bell curve for the one-dimensional or\n        univariate normal distribution.\n\n        Covariance indicates the level to which two variables vary together.\n        From the multivariate normal distribution, we"" draw N-dimensional\n        samples, :math:`X = [x_1, x_2, ... x_N]`.  The covariance matrix\n        element :math:`C_{ij}` is the covariance of :math:`x_i` and :math:`x_j`.\n        The element :math:`C_{ii}` is the variance of :math:`x_i` (i.e. its\n        \"spread\").\n\n        Instead of specifying the full covariance matrix, popular\n        approximations include:\n\n          - Spherical covariance (*cov* is a multiple of the identity matrix)\n          - Diagonal covariance (*cov* has non-negative elements, and only on\n            the diagonal)\n\n        This geometrical property can be seen in two dimensions by plotting\n        generated data-points:\n\n        >>> mean = [0,0]\n        >>> cov = [[1,0],[0,100]] # diagonal covariance, points lie on x or y-axis\n\n        >>> import matplotlib.pyplot as plt\n        >>> x,y = np.random.multivariate_normal(mean,cov,5000).T\n        >>> plt.plot(x,y,'x'); plt.axis('equal'); plt.show()\n\n        Note that the covariance matrix must be non-negative definite.\n\n        References\n        ----------\n        Papoulis, A., *Probability, Random Variables, and Stochastic Processes*,\n        3rd ed., New York: McGraw-Hill, 1991.\n\n        Duda, R. O., Hart, P. E., and Stork, D. G., *Pattern Classification*,\n        2nd ed., New York: Wiley, 2001.\n\n        Examples\n        --------\n        >>> mean = (1,2)\n        >>> cov = [[1,0],[1,0]]\n        >>> x = np.random.multivariate_normal(mean,cov,(3,3))\n        >>> x.shape\n        (3, 3, 2)\n\n        The following is probably true, given that 0.6 is roughly twice the\n        standard deviation:\n\n        >>> print list( (x[0,0,:] - mean) < 0.6 )\n        [True, True]\n\n        ";
-static char __pyx_k_284[] = "RandomState.multinomial (line 4205)";
-static char __pyx_k_285[] = "\n        multinomial(n, pvals, size=None)\n\n        Draw samples from a multinomial distribution.\n\n        The multinomial distribution is a multivariate generalisation of the\n        binomial distribution.  Take an experiment with one of ``p``\n        possible outcomes.  An example of such an experiment is throwing a dice,\n        where the outcome can be 1 through 6.  Each sample drawn from the\n        distribution represents `n` such experiments.  Its values,\n        ``X_i = [X_0, X_1, ..., X_p]``, represent the number of times the outcome\n        was ``i``.\n\n        Parameters\n        ----------\n        n : int\n            Number of experiments.\n        pvals : sequence of floats, length p\n            Probabilities of each of the ``p`` different outcomes.  These\n            should sum to 1 (however, the last element is always assumed to\n            account for the remaining probability, as long as\n            ``sum(pvals[:-1]) <= 1)``.\n        size : tuple of ints\n            Given a `size` of ``(M, N, K)``, then ``M*N*K`` samples are drawn,\n            and the output shape becomes ``(M, N, K, p)``, since each sample\n            has shape ``(p,)``.\n\n        Examples\n        --------\n        Throw a dice 20 times:\n\n        >>> np.random.multinomial(20, [1/6.]*6, size=1)\n        array([[4, 1, 7, 5, 2, 1]])\n\n        It landed 4 times on 1, once on 2, etc.\n\n        Now, throw the dice 20 times, and 20 times again:\n\n        >>> np.random.multinomial(20, [1/6.]*6, size=2)\n        array([[3, 4, 3, 3, 4, 3],\n               [2, 4, 3, 4, 0, 7]])\n\n        For the first run, we threw 3 times 1, 4 times 2, etc.  For the second,\n        we threw 2 times 1, 4 times 2, etc.\n\n        A loaded dice is more likely to land on number 6:\n\n        >>> np.random.multinomial(100, [1/7.]*5)\n        array([13, 16, 13, 16, 42])\n\n        ";
-static char __pyx_k_286[] = "RandomState.dirichlet (line 4293)";
-static char __pyx_k_287[] = "\n        dirichlet(alpha, size=None)\n\n        Draw samples from the Dirichlet distribution.\n\n        Draw `size` samples of dimension k from a Dirichlet distribution. A\n        Dirichlet-distributed random variable can be seen as a multivariate\n        generalization of a Beta distribution. Dirichlet pdf is the conjugate\n        prior of a multinomial in Bayesian inference.\n\n        Parameters\n        ----------\n        alpha : array\n            Parameter of the distribution (k dimension for sample of\n            dimension k).\n        size : array\n            Number of samples to draw.\n\n        Returns\n        -------\n        samples : ndarray,\n            The drawn samples, of shape (alpha.ndim, size).\n\n        Notes\n        -----\n        .. math:: X \\approx \\prod_{i=1}^{k}{x^{\\alpha_i-1}_i}\n\n        Uses the following property for computation: for each dimension,\n        draw a random sample y_i from a standard gamma generator of shape\n        `alpha_i`, then\n        :math:`X = \\frac{1}{\\sum_{i=1}^k{y_i}} (y_1, \\ldots, y_n)` is\n        Dirichlet distributed.\n\n        References\n        ----------\n        .. [1] David McKay, \"Information Theory, Inference and Learning\n               Algorithms,\" chapter 23,\n               http://www.inference.phy.cam.ac.uk/mackay/\n        .. [2] Wikipedia, \"Dirichlet distribution\",\n               http://en.wikipedia.org/wiki/Dirichlet_distribution\n\n        Examples\n        --------\n        Taking an example cited in Wikipedia, this distribution can be used if\n        one wanted to cut strings (each of initial length 1.0) into K pieces\n        with different lengths, where each piece had, on average, a designated\n        average length, but allowing some variation in the relative sizes of the\n        pieces.\n\n        >>> s = np.random.dirichlet((10, 5, 3), 20).transpose()\n\n        >>> plt.barh(range(20), s[0])\n        >>> plt.barh(range(20), s[1], left=s[0], color='g')""\n        >>> plt.barh(range(20), s[2], left=s[0]+s[1], color='r')\n        >>> plt.title(\"Lengths of Strings\")\n\n        ";
-static char __pyx_k_288[] = "RandomState.shuffle (line 4405)";
-static char __pyx_k_289[] = "\n        shuffle(x)\n\n        Modify a sequence in-place by shuffling its contents.\n\n        Parameters\n        ----------\n        x : array_like\n            The array or list to be shuffled.\n\n        Returns\n        -------\n        None\n\n        Examples\n        --------\n        >>> arr = np.arange(10)\n        >>> np.random.shuffle(arr)\n        >>> arr\n        [1 7 5 2 9 4 3 6 0 8]\n\n        This function only shuffles the array along the first index of a\n        multi-dimensional array:\n\n        >>> arr = np.arange(9).reshape((3, 3))\n        >>> np.random.shuffle(arr)\n        >>> arr\n        array([[3, 4, 5],\n               [6, 7, 8],\n               [0, 1, 2]])\n\n        ";
-static char __pyx_k_290[] = "RandomState.permutation (line 4464)";
-static char __pyx_k_291[] = "\n        permutation(x)\n\n        Randomly permute a sequence, or return a permuted range.\n\n        If `x` is a multi-dimensional array, it is only shuffled along its\n        first index.\n\n        Parameters\n        ----------\n        x : int or array_like\n            If `x` is an integer, randomly permute ``np.arange(x)``.\n            If `x` is an array, make a copy and shuffle the elements\n            randomly.\n\n        Returns\n        -------\n        out : ndarray\n            Permuted sequence or array range.\n\n        Examples\n        --------\n        >>> np.random.permutation(10)\n        array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6])\n\n        >>> np.random.permutation([1, 4, 9, 12, 15])\n        array([15,  1,  9,  4, 12])\n\n        >>> arr = np.arange(9).reshape((3, 3))\n        >>> np.random.permutation(arr)\n        array([[6, 7, 8],\n               [0, 1, 2],\n               [3, 4, 5]])\n\n        ";
+static char __pyx_k_141[] = "p is nan";
+static char __pyx_k_146[] = "n <= 0";
+static char __pyx_k_154[] = "lam < 0";
+static char __pyx_k_156[] = "lam value too large";
+static char __pyx_k_159[] = "lam value too large.";
+static char __pyx_k_161[] = "a <= 1.0";
+static char __pyx_k_164[] = "p < 0.0";
+static char __pyx_k_166[] = "p > 1.0";
+static char __pyx_k_170[] = "ngood < 0";
+static char __pyx_k_172[] = "nbad < 0";
+static char __pyx_k_174[] = "nsample < 1";
+static char __pyx_k_176[] = "ngood + nbad < nsample";
+static char __pyx_k_182[] = "p <= 0.0";
+static char __pyx_k_184[] = "p >= 1.0";
+static char __pyx_k_188[] = "numpy.dual";
+static char __pyx_k_189[] = "mean must be 1 dimensional";
+static char __pyx_k_191[] = "cov must be 2 dimensional and square";
+static char __pyx_k_193[] = "mean and cov must have same length";
+static char __pyx_k_196[] = "covariance is not positive-semidefinite.";
+static char __pyx_k_199[] = "sum(pvals[:-1]) > 1.0";
+static char __pyx_k_203[] = "/home/daniel/repos/numpy/numpy/random/mtrand/mtrand.pyx";
+static char __pyx_k_206[] = "standard_exponential";
+static char __pyx_k_207[] = "noncentral_chisquare";
+static char __pyx_k_208[] = "RandomState.random_sample (line 731)";
+static char __pyx_k_209[] = "\n        random_sample(size=None)\n\n        Return random floats in the half-open interval [0.0, 1.0).\n\n        Results are from the \"continuous uniform\" distribution over the\n        stated interval.  To sample :math:`Unif[a, b), b > a` multiply\n        the output of `random_sample` by `(b-a)` and add `a`::\n\n          (b - a) * random_sample() + a\n\n        Parameters\n        ----------\n        size : int or tuple of ints, optional\n            Defines the shape of the returned array of random floats. If None\n            (the default), returns a single float.\n\n        Returns\n        -------\n        out : float or ndarray of floats\n            Array of random floats of shape `size` (unless ``size=None``, in which\n            case a single float is returned).\n\n        Examples\n        --------\n        >>> np.random.random_sample()\n        0.47108547995356098\n        >>> type(np.random.random_sample())\n        <type 'float'>\n        >>> np.random.random_sample((5,))\n        array([ 0.30220482,  0.86820401,  0.1654503 ,  0.11659149,  0.54323428])\n\n        Three-by-two array of random numbers from [-5, 0):\n\n        >>> 5 * np.random.random_sample((3, 2)) - 5\n        array([[-3.99149989, -0.52338984],\n               [-2.99091858, -0.79479508],\n               [-1.23204345, -1.75224494]])\n\n        ";
+static char __pyx_k_210[] = "RandomState.tomaxint (line 774)";
+static char __pyx_k_211[] = "\n        tomaxint(size=None)\n\n        Random integers between 0 and ``sys.maxint``, inclusive.\n\n        Return a sample of uniformly distributed random integers in the interval\n        [0, ``sys.maxint``].\n\n        Parameters\n        ----------\n        size : tuple of ints, int, optional\n            Shape of output.  If this is, for example, (m,n,k), m*n*k samples\n            are generated.  If no shape is specified, a single sample is\n            returned.\n\n        Returns\n        -------\n        out : ndarray\n            Drawn samples, with shape `size`.\n\n        See Also\n        --------\n        randint : Uniform sampling over a given half-open interval of integers.\n        random_integers : Uniform sampling over a given closed interval of\n            integers.\n\n        Examples\n        --------\n        >>> RS = np.random.mtrand.RandomState() # need a RandomState object\n        >>> RS.tomaxint((2,2,2))\n        array([[[1170048599, 1600360186],\n                [ 739731006, 1947757578]],\n               [[1871712945,  752307660],\n                [1601631370, 1479324245]]])\n        >>> import sys\n        >>> sys.maxint\n        2147483647\n        >>> RS.tomaxint((2,2,2)) < sys.maxint\n        array([[[ True,  True],\n                [ True,  True]],\n               [[ True,  True],\n                [ True,  True]]], dtype=bool)\n\n        ";
+static char __pyx_k_212[] = "RandomState.randint (line 821)";
+static char __pyx_k_213[] = "\n        randint(low, high=None, size=None)\n\n        Return random integers from `low` (inclusive) to `high` (exclusive).\n\n        Return random integers from the \"discrete uniform\" distribution in the\n        \"half-open\" interval [`low`, `high`). If `high` is None (the default),\n        then results are from [0, `low`).\n\n        Parameters\n        ----------\n        low : int\n            Lowest (signed) integer to be drawn from the distribution (unless\n            ``high=None``, in which case this parameter is the *highest* such\n            integer).\n        high : int, optional\n            If provided, one above the largest (signed) integer to be drawn\n            from the distribution (see above for behavior if ``high=None``).\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single int is\n            returned.\n\n        Returns\n        -------\n        out : int or ndarray of ints\n            `size`-shaped array of random integers from the appropriate\n            distribution, or a single such random int if `size` not provided.\n\n        See Also\n        --------\n        random.random_integers : similar to `randint`, only for the closed\n            interval [`low`, `high`], and 1 is the lowest value if `high` is\n            omitted. In particular, this other one is the one to use to generate\n            uniformly distributed discrete non-integers.\n\n        Examples\n        --------\n        >>> np.random.randint(2, size=10)\n        array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0])\n        >>> np.random.randint(1, size=10)\n        array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0])\n\n        Generate a 2 x 4 array of ints between 0 and 4, inclusive:\n\n        >>> np.random.randint(5, size=(2, 4))\n        array([[4, 0, 2, 1],\n               [3, 2, 2, 0]])\n\n        ";
+static char __pyx_k_214[] = "RandomState.bytes (line 901)";
+static char __pyx_k_215[] = "\n        bytes(length)\n\n        Return random bytes.\n\n        Parameters\n        ----------\n        length : int\n            Number of random bytes.\n\n        Returns\n        -------\n        out : str\n            String of length `length`.\n\n        Examples\n        --------\n        >>> np.random.bytes(10)\n        ' eh\\x85\\x022SZ\\xbf\\xa4' #random\n\n        ";
+static char __pyx_k_216[] = "RandomState.choice (line 929)";
+static char __pyx_k_217[] = "\n        choice(a, size=None, replace=True, p=None)\n\n        Generates a random sample from a given 1-D array\n\n                .. versionadded:: 1.7.0\n\n        Parameters\n        -----------\n        a : 1-D array-like or int\n            If an ndarray, a random sample is generated from its elements.\n            If an int, the random sample is generated as if a was np.arange(n)\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n        replace : boolean, optional\n            Whether the sample is with or without replacement\n        p : 1-D array-like, optional\n            The probabilities associated with each entry in a.\n            If not given the sample assumes a uniform distribtion over all\n            entries in a.\n\n        Returns\n        --------\n        samples : 1-D ndarray, shape (size,)\n            The generated random samples\n\n        Raises\n        -------\n        ValueError\n            If a is an int and less than zero, if a or p are not 1-dimensional,\n            if a is an array-like of size 0, if p is not a vector of\n            probabilities, if a and p have different lengths, or if\n            replace=False and the sample size is greater than the population\n            size\n\n        See Also\n        ---------\n        randint, shuffle, permutation\n\n        Examples\n        ---------\n        Generate a uniform random sample from np.arange(5) of size 3:\n\n        >>> np.random.choice(5, 3)\n        array([0, 3, 4])\n        >>> #This is equivalent to np.random.randint(0,5,3)\n\n        Generate a non-uniform random sample from np.arange(5) of size 3:\n\n        >>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0])\n        array([3, 3, 0])\n\n        Generate a uniform random sample from np.arange(5) of size 3 without\n        replacement:\n\n        >>> np.random.choice(5, 3, replace=False)\n        array([3,1,0])\n    ""    >>> #This is equivalent to np.random.permutation(np.arange(5))[:3]\n\n        Generate a non-uniform random sample from np.arange(5) of size\n        3 without replacement:\n\n        >>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0])\n        array([2, 3, 0])\n\n        Any of the above can be repeated with an arbitrary array-like\n        instead of just integers. For instance:\n\n        >>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher']\n        >>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3])\n        array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'],\n              dtype='|S11')\n\n        ";
+static char __pyx_k_218[] = "RandomState.uniform (line 1104)";
+static char __pyx_k_219[] = "\n        uniform(low=0.0, high=1.0, size=1)\n\n        Draw samples from a uniform distribution.\n\n        Samples are uniformly distributed over the half-open interval\n        ``[low, high)`` (includes low, but excludes high).  In other words,\n        any value within the given interval is equally likely to be drawn\n        by `uniform`.\n\n        Parameters\n        ----------\n        low : float, optional\n            Lower boundary of the output interval.  All values generated will be\n            greater than or equal to low.  The default value is 0.\n        high : float\n            Upper boundary of the output interval.  All values generated will be\n            less than high.  The default value is 1.0.\n        size : int or tuple of ints, optional\n            Shape of output.  If the given size is, for example, (m,n,k),\n            m*n*k samples are generated.  If no shape is specified, a single sample\n            is returned.\n\n        Returns\n        -------\n        out : ndarray\n            Drawn samples, with shape `size`.\n\n        See Also\n        --------\n        randint : Discrete uniform distribution, yielding integers.\n        random_integers : Discrete uniform distribution over the closed\n                          interval ``[low, high]``.\n        random_sample : Floats uniformly distributed over ``[0, 1)``.\n        random : Alias for `random_sample`.\n        rand : Convenience function that accepts dimensions as input, e.g.,\n               ``rand(2,2)`` would generate a 2-by-2 array of floats,\n               uniformly distributed over ``[0, 1)``.\n\n        Notes\n        -----\n        The probability density function of the uniform distribution is\n\n        .. math:: p(x) = \\frac{1}{b - a}\n\n        anywhere within the interval ``[a, b)``, and zero elsewhere.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> s = np.random.uniform(-1,0,1000)\n\n        All values are w""ithin the given interval:\n\n        >>> np.all(s >= -1)\n        True\n        >>> np.all(s < 0)\n        True\n\n        Display the histogram of the samples, along with the\n        probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 15, normed=True)\n        >>> plt.plot(bins, np.ones_like(bins), linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_220[] = "RandomState.rand (line 1191)";
+static char __pyx_k_221[] = "\n        rand(d0, d1, ..., dn)\n\n        Random values in a given shape.\n\n        Create an array of the given shape and propagate it with\n        random samples from a uniform distribution\n        over ``[0, 1)``.\n\n        Parameters\n        ----------\n        d0, d1, ..., dn : int, optional\n            The dimensions of the returned array, should all be positive.\n            If no argument is given a single Python float is returned.\n\n        Returns\n        -------\n        out : ndarray, shape ``(d0, d1, ..., dn)``\n            Random values.\n\n        See Also\n        --------\n        random\n\n        Notes\n        -----\n        This is a convenience function. If you want an interface that\n        takes a shape-tuple as the first argument, refer to\n        np.random.random_sample .\n\n        Examples\n        --------\n        >>> np.random.rand(3,2)\n        array([[ 0.14022471,  0.96360618],  #random\n               [ 0.37601032,  0.25528411],  #random\n               [ 0.49313049,  0.94909878]]) #random\n\n        ";
+static char __pyx_k_222[] = "RandomState.randn (line 1235)";
+static char __pyx_k_223[] = "\n        randn(d0, d1, ..., dn)\n\n        Return a sample (or samples) from the \"standard normal\" distribution.\n\n        If positive, int_like or int-convertible arguments are provided,\n        `randn` generates an array of shape ``(d0, d1, ..., dn)``, filled\n        with random floats sampled from a univariate \"normal\" (Gaussian)\n        distribution of mean 0 and variance 1 (if any of the :math:`d_i` are\n        floats, they are first converted to integers by truncation). A single\n        float randomly sampled from the distribution is returned if no\n        argument is provided.\n\n        This is a convenience function.  If you want an interface that takes a\n        tuple as the first argument, use `numpy.random.standard_normal` instead.\n\n        Parameters\n        ----------\n        d0, d1, ..., dn : int, optional\n            The dimensions of the returned array, should be all positive.\n            If no argument is given a single Python float is returned.\n\n        Returns\n        -------\n        Z : ndarray or float\n            A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from\n            the standard normal distribution, or a single such float if\n            no parameters were supplied.\n\n        See Also\n        --------\n        random.standard_normal : Similar, but takes a tuple as its argument.\n\n        Notes\n        -----\n        For random samples from :math:`N(\\mu, \\sigma^2)`, use:\n\n        ``sigma * np.random.randn(...) + mu``\n\n        Examples\n        --------\n        >>> np.random.randn()\n        2.1923875335537315 #random\n\n        Two-by-four array of samples from N(3, 6.25):\n\n        >>> 2.5 * np.random.randn(2, 4) + 3\n        array([[-4.49401501,  4.00950034, -1.81814867,  7.29718677],  #random\n               [ 0.39924804,  4.68456316,  4.99394529,  4.84057254]]) #random\n\n        ";
+static char __pyx_k_224[] = "RandomState.random_integers (line 1292)";
+static char __pyx_k_225[] = "\n        random_integers(low, high=None, size=None)\n\n        Return random integers between `low` and `high`, inclusive.\n\n        Return random integers from the \"discrete uniform\" distribution in the\n        closed interval [`low`, `high`].  If `high` is None (the default),\n        then results are from [1, `low`].\n\n        Parameters\n        ----------\n        low : int\n            Lowest (signed) integer to be drawn from the distribution (unless\n            ``high=None``, in which case this parameter is the *highest* such\n            integer).\n        high : int, optional\n            If provided, the largest (signed) integer to be drawn from the\n            distribution (see above for behavior if ``high=None``).\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single int is returned.\n\n        Returns\n        -------\n        out : int or ndarray of ints\n            `size`-shaped array of random integers from the appropriate\n            distribution, or a single such random int if `size` not provided.\n\n        See Also\n        --------\n        random.randint : Similar to `random_integers`, only for the half-open\n            interval [`low`, `high`), and 0 is the lowest value if `high` is\n            omitted.\n\n        Notes\n        -----\n        To sample from N evenly spaced floating-point numbers between a and b,\n        use::\n\n          a + (b - a) * (np.random.random_integers(N) - 1) / (N - 1.)\n\n        Examples\n        --------\n        >>> np.random.random_integers(5)\n        4\n        >>> type(np.random.random_integers(5))\n        <type 'int'>\n        >>> np.random.random_integers(5, size=(3.,2.))\n        array([[5, 4],\n               [3, 3],\n               [4, 5]])\n\n        Choose five random numbers from the set of five evenly-spaced\n        numbers between 0 and 2.5, inclusive (*i.e.*, from the set\n        :math:`{0, 5/8, 10/8, 15/8, 20/8}`):\n""\n        >>> 2.5 * (np.random.random_integers(5, size=(5,)) - 1) / 4.\n        array([ 0.625,  1.25 ,  0.625,  0.625,  2.5  ])\n\n        Roll two six sided dice 1000 times and sum the results:\n\n        >>> d1 = np.random.random_integers(1, 6, 1000)\n        >>> d2 = np.random.random_integers(1, 6, 1000)\n        >>> dsums = d1 + d2\n\n        Display results as a histogram:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(dsums, 11, normed=True)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_226[] = "RandomState.standard_normal (line 1370)";
+static char __pyx_k_227[] = "\n        standard_normal(size=None)\n\n        Returns samples from a Standard Normal distribution (mean=0, stdev=1).\n\n        Parameters\n        ----------\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        out : float or ndarray\n            Drawn samples.\n\n        Examples\n        --------\n        >>> s = np.random.standard_normal(8000)\n        >>> s\n        array([ 0.6888893 ,  0.78096262, -0.89086505, ...,  0.49876311, #random\n               -0.38672696, -0.4685006 ])                               #random\n        >>> s.shape\n        (8000,)\n        >>> s = np.random.standard_normal(size=(3, 4, 2))\n        >>> s.shape\n        (3, 4, 2)\n\n        ";
+static char __pyx_k_228[] = "RandomState.normal (line 1402)";
+static char __pyx_k_229[] = "\n        normal(loc=0.0, scale=1.0, size=None)\n\n        Draw random samples from a normal (Gaussian) distribution.\n\n        The probability density function of the normal distribution, first\n        derived by De Moivre and 200 years later by both Gauss and Laplace\n        independently [2]_, is often called the bell curve because of\n        its characteristic shape (see the example below).\n\n        The normal distributions occurs often in nature.  For example, it\n        describes the commonly occurring distribution of samples influenced\n        by a large number of tiny, random disturbances, each with its own\n        unique distribution [2]_.\n\n        Parameters\n        ----------\n        loc : float\n            Mean (\"centre\") of the distribution.\n        scale : float\n            Standard deviation (spread or \"width\") of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        See Also\n        --------\n        scipy.stats.distributions.norm : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Gaussian distribution is\n\n        .. math:: p(x) = \\frac{1}{\\sqrt{ 2 \\pi \\sigma^2 }}\n                         e^{ - \\frac{ (x - \\mu)^2 } {2 \\sigma^2} },\n\n        where :math:`\\mu` is the mean and :math:`\\sigma` the standard deviation.\n        The square of the standard deviation, :math:`\\sigma^2`, is called the\n        variance.\n\n        The function has its peak at the mean, and its \"spread\" increases with\n        the standard deviation (the function reaches 0.607 times its maximum at\n        :math:`x + \\sigma` and :math:`x - \\sigma` [2]_).  This implies that\n        `numpy.random.normal` is more likely to return samples lying close to the\n        mean, rather than those far away.\n""\n        References\n        ----------\n        .. [1] Wikipedia, \"Normal distribution\",\n               http://en.wikipedia.org/wiki/Normal_distribution\n        .. [2] P. R. Peebles Jr., \"Central Limit Theorem\" in \"Probability, Random\n               Variables and Random Signal Principles\", 4th ed., 2001,\n               pp. 51, 51, 125.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, sigma = 0, 0.1 # mean and standard deviation\n        >>> s = np.random.normal(mu, sigma, 1000)\n\n        Verify the mean and the variance:\n\n        >>> abs(mu - np.mean(s)) < 0.01\n        True\n\n        >>> abs(sigma - np.std(s, ddof=1)) < 0.01\n        True\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 30, normed=True)\n        >>> plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) *\n        ...                np.exp( - (bins - mu)**2 / (2 * sigma**2) ),\n        ...          linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_230[] = "RandomState.standard_exponential (line 1615)";
+static char __pyx_k_231[] = "\n        standard_exponential(size=None)\n\n        Draw samples from the standard exponential distribution.\n\n        `standard_exponential` is identical to the exponential distribution\n        with a scale parameter of 1.\n\n        Parameters\n        ----------\n        size : int or tuple of ints\n            Shape of the output.\n\n        Returns\n        -------\n        out : float or ndarray\n            Drawn samples.\n\n        Examples\n        --------\n        Output a 3x8000 array:\n\n        >>> n = np.random.standard_exponential((3, 8000))\n\n        ";
+static char __pyx_k_232[] = "RandomState.standard_gamma (line 1643)";
+static char __pyx_k_233[] = "\n        standard_gamma(shape, size=None)\n\n        Draw samples from a Standard Gamma distribution.\n\n        Samples are drawn from a Gamma distribution with specified parameters,\n        shape (sometimes designated \"k\") and scale=1.\n\n        Parameters\n        ----------\n        shape : float\n            Parameter, should be > 0.\n        size : int or tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The drawn samples.\n\n        See Also\n        --------\n        scipy.stats.distributions.gamma : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Gamma distribution is\n\n        .. math:: p(x) = x^{k-1}\\frac{e^{-x/\\theta}}{\\theta^k\\Gamma(k)},\n\n        where :math:`k` is the shape and :math:`\\theta` the scale,\n        and :math:`\\Gamma` is the Gamma function.\n\n        The Gamma distribution is often used to model the times to failure of\n        electronic components, and arises naturally in processes for which the\n        waiting times between Poisson distributed events are relevant.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Gamma Distribution.\" From MathWorld--A\n               Wolfram Web Resource.\n               http://mathworld.wolfram.com/GammaDistribution.html\n        .. [2] Wikipedia, \"Gamma-distribution\",\n               http://en.wikipedia.org/wiki/Gamma-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> shape, scale = 2., 1. # mean and width\n        >>> s = np.random.standard_gamma(shape, 1000000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt""\n        >>> import scipy.special as sps\n        >>> count, bins, ignored = plt.hist(s, 50, normed=True)\n        >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ \\\n        ...                       (sps.gamma(shape) * scale**shape))\n        >>> plt.plot(bins, y, linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_234[] = "RandomState.gamma (line 1725)";
+static char __pyx_k_235[] = "\n        gamma(shape, scale=1.0, size=None)\n\n        Draw samples from a Gamma distribution.\n\n        Samples are drawn from a Gamma distribution with specified parameters,\n        `shape` (sometimes designated \"k\") and `scale` (sometimes designated\n        \"theta\"), where both parameters are > 0.\n\n        Parameters\n        ----------\n        shape : scalar > 0\n            The shape of the gamma distribution.\n        scale : scalar > 0, optional\n            The scale of the gamma distribution.  Default is equal to 1.\n        size : shape_tuple, optional\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        out : ndarray, float\n            Returns one sample unless `size` parameter is specified.\n\n        See Also\n        --------\n        scipy.stats.distributions.gamma : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Gamma distribution is\n\n        .. math:: p(x) = x^{k-1}\\frac{e^{-x/\\theta}}{\\theta^k\\Gamma(k)},\n\n        where :math:`k` is the shape and :math:`\\theta` the scale,\n        and :math:`\\Gamma` is the Gamma function.\n\n        The Gamma distribution is often used to model the times to failure of\n        electronic components, and arises naturally in processes for which the\n        waiting times between Poisson distributed events are relevant.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Gamma Distribution.\" From MathWorld--A\n               Wolfram Web Resource.\n               http://mathworld.wolfram.com/GammaDistribution.html\n        .. [2] Wikipedia, \"Gamma-distribution\",\n               http://en.wikipedia.org/wiki/Gamma-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> shape, scale = 2.,"" 2. # mean and dispersion\n        >>> s = np.random.gamma(shape, scale, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> import scipy.special as sps\n        >>> count, bins, ignored = plt.hist(s, 50, normed=True)\n        >>> y = bins**(shape-1)*(np.exp(-bins/scale) /\n        ...                      (sps.gamma(shape)*scale**shape))\n        >>> plt.plot(bins, y, linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_236[] = "RandomState.f (line 1816)";
+static char __pyx_k_237[] = "\n        f(dfnum, dfden, size=None)\n\n        Draw samples from a F distribution.\n\n        Samples are drawn from an F distribution with specified parameters,\n        `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom\n        in denominator), where both parameters should be greater than zero.\n\n        The random variate of the F distribution (also known as the\n        Fisher distribution) is a continuous probability distribution\n        that arises in ANOVA tests, and is the ratio of two chi-square\n        variates.\n\n        Parameters\n        ----------\n        dfnum : float\n            Degrees of freedom in numerator. Should be greater than zero.\n        dfden : float\n            Degrees of freedom in denominator. Should be greater than zero.\n        size : {tuple, int}, optional\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``,\n            then ``m * n * k`` samples are drawn. By default only one sample\n            is returned.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n            Samples from the Fisher distribution.\n\n        See Also\n        --------\n        scipy.stats.distributions.f : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The F statistic is used to compare in-group variances to between-group\n        variances. Calculating the distribution depends on the sampling, and\n        so it is a function of the respective degrees of freedom in the\n        problem.  The variable `dfnum` is the number of samples minus one, the\n        between-groups degrees of freedom, while `dfden` is the within-groups\n        degrees of freedom, the sum of the number of samples in each group\n        minus the number of groups.\n\n        References\n        ----------\n        .. [1] Glantz, Stanton A. \"Primer of Biostatistics.\", McGraw-Hill,\n               Fifth Edition, 2002.""\n        .. [2] Wikipedia, \"F-distribution\",\n               http://en.wikipedia.org/wiki/F-distribution\n\n        Examples\n        --------\n        An example from Glantz[1], pp 47-40.\n        Two groups, children of diabetics (25 people) and children from people\n        without diabetes (25 controls). Fasting blood glucose was measured,\n        case group had a mean value of 86.1, controls had a mean value of\n        82.2. Standard deviations were 2.09 and 2.49 respectively. Are these\n        data consistent with the null hypothesis that the parents diabetic\n        status does not affect their children's blood glucose levels?\n        Calculating the F statistic from the data gives a value of 36.01.\n\n        Draw samples from the distribution:\n\n        >>> dfnum = 1. # between group degrees of freedom\n        >>> dfden = 48. # within groups degrees of freedom\n        >>> s = np.random.f(dfnum, dfden, 1000)\n\n        The lower bound for the top 1% of the samples is :\n\n        >>> sort(s)[-10]\n        7.61988120985\n\n        So there is about a 1% chance that the F statistic will exceed 7.62,\n        the measured value is 36, so the null hypothesis is rejected at the 1%\n        level.\n\n        ";
+static char __pyx_k_238[] = "RandomState.noncentral_f (line 1918)";
+static char __pyx_k_239[] = "\n        noncentral_f(dfnum, dfden, nonc, size=None)\n\n        Draw samples from the noncentral F distribution.\n\n        Samples are drawn from an F distribution with specified parameters,\n        `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of\n        freedom in denominator), where both parameters > 1.\n        `nonc` is the non-centrality parameter.\n\n        Parameters\n        ----------\n        dfnum : int\n            Parameter, should be > 1.\n        dfden : int\n            Parameter, should be > 1.\n        nonc : float\n            Parameter, should be >= 0.\n        size : int or tuple of ints\n            Output shape. If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : scalar or ndarray\n            Drawn samples.\n\n        Notes\n        -----\n        When calculating the power of an experiment (power = probability of\n        rejecting the null hypothesis when a specific alternative is true) the\n        non-central F statistic becomes important.  When the null hypothesis is\n        true, the F statistic follows a central F distribution. When the null\n        hypothesis is not true, then it follows a non-central F statistic.\n\n        References\n        ----------\n        Weisstein, Eric W. \"Noncentral F-Distribution.\" From MathWorld--A Wolfram\n        Web Resource.  http://mathworld.wolfram.com/NoncentralF-Distribution.html\n\n        Wikipedia, \"Noncentral F distribution\",\n        http://en.wikipedia.org/wiki/Noncentral_F-distribution\n\n        Examples\n        --------\n        In a study, testing for a specific alternative to the null hypothesis\n        requires use of the Noncentral F distribution. We need to calculate the\n        area in the tail of the distribution that exceeds the value of the F\n        distribution for the null hypothesis.  We'll plot the two probability\n        distributions for comp""arison.\n\n        >>> dfnum = 3 # between group deg of freedom\n        >>> dfden = 20 # within groups degrees of freedom\n        >>> nonc = 3.0\n        >>> nc_vals = np.random.noncentral_f(dfnum, dfden, nonc, 1000000)\n        >>> NF = np.histogram(nc_vals, bins=50, normed=True)\n        >>> c_vals = np.random.f(dfnum, dfden, 1000000)\n        >>> F = np.histogram(c_vals, bins=50, normed=True)\n        >>> plt.plot(F[1][1:], F[0])\n        >>> plt.plot(NF[1][1:], NF[0])\n        >>> plt.show()\n\n        ";
+static char __pyx_k_240[] = "RandomState.chisquare (line 2013)";
+static char __pyx_k_241[] = "\n        chisquare(df, size=None)\n\n        Draw samples from a chi-square distribution.\n\n        When `df` independent random variables, each with standard normal\n        distributions (mean 0, variance 1), are squared and summed, the\n        resulting distribution is chi-square (see Notes).  This distribution\n        is often used in hypothesis testing.\n\n        Parameters\n        ----------\n        df : int\n             Number of degrees of freedom.\n        size : tuple of ints, int, optional\n             Size of the returned array.  By default, a scalar is\n             returned.\n\n        Returns\n        -------\n        output : ndarray\n            Samples drawn from the distribution, packed in a `size`-shaped\n            array.\n\n        Raises\n        ------\n        ValueError\n            When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``)\n            is given.\n\n        Notes\n        -----\n        The variable obtained by summing the squares of `df` independent,\n        standard normally distributed random variables:\n\n        .. math:: Q = \\sum_{i=0}^{\\mathtt{df}} X^2_i\n\n        is chi-square distributed, denoted\n\n        .. math:: Q \\sim \\chi^2_k.\n\n        The probability density function of the chi-squared distribution is\n\n        .. math:: p(x) = \\frac{(1/2)^{k/2}}{\\Gamma(k/2)}\n                         x^{k/2 - 1} e^{-x/2},\n\n        where :math:`\\Gamma` is the gamma function,\n\n        .. math:: \\Gamma(x) = \\int_0^{-\\infty} t^{x - 1} e^{-t} dt.\n\n        References\n        ----------\n        `NIST/SEMATECH e-Handbook of Statistical Methods\n        <http://www.itl.nist.gov/div898/handbook/eda/section3/eda3666.htm>`_\n\n        Examples\n        --------\n        >>> np.random.chisquare(2,4)\n        array([ 1.89920014,  9.00867716,  3.13710533,  5.62318272])\n\n        ";
+static char __pyx_k_242[] = "RandomState.noncentral_chisquare (line 2091)";
+static char __pyx_k_243[] = "\n        noncentral_chisquare(df, nonc, size=None)\n\n        Draw samples from a noncentral chi-square distribution.\n\n        The noncentral :math:`\\chi^2` distribution is a generalisation of\n        the :math:`\\chi^2` distribution.\n\n        Parameters\n        ----------\n        df : int\n            Degrees of freedom, should be >= 1.\n        nonc : float\n            Non-centrality, should be > 0.\n        size : int or tuple of ints\n            Shape of the output.\n\n        Notes\n        -----\n        The probability density function for the noncentral Chi-square distribution\n        is\n\n        .. math:: P(x;df,nonc) = \\sum^{\\infty}_{i=0}\n                               \\frac{e^{-nonc/2}(nonc/2)^{i}}{i!}P_{Y_{df+2i}}(x),\n\n        where :math:`Y_{q}` is the Chi-square with q degrees of freedom.\n\n        In Delhi (2007), it is noted that the noncentral chi-square is useful in\n        bombing and coverage problems, the probability of killing the point target\n        given by the noncentral chi-squared distribution.\n\n        References\n        ----------\n        .. [1] Delhi, M.S. Holla, \"On a noncentral chi-square distribution in the\n               analysis of weapon systems effectiveness\", Metrika, Volume 15,\n               Number 1 / December, 1970.\n        .. [2] Wikipedia, \"Noncentral chi-square distribution\"\n               http://en.wikipedia.org/wiki/Noncentral_chi-square_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram\n\n        >>> import matplotlib.pyplot as plt\n        >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000),\n        ...                   bins=200, normed=True)\n        >>> plt.show()\n\n        Draw values from a noncentral chisquare with very small noncentrality,\n        and compare to a chisquare.\n\n        >>> plt.figure()\n        >>> values = plt.hist(np.random.noncentral_chisquare(3, .0000001, 100000),\n     ""   ...                   bins=np.arange(0., 25, .1), normed=True)\n        >>> values2 = plt.hist(np.random.chisquare(3, 100000),\n        ...                    bins=np.arange(0., 25, .1), normed=True)\n        >>> plt.plot(values[1][0:-1], values[0]-values2[0], 'ob')\n        >>> plt.show()\n\n        Demonstrate how large values of non-centrality lead to a more symmetric\n        distribution.\n\n        >>> plt.figure()\n        >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000),\n        ...                   bins=200, normed=True)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_244[] = "RandomState.standard_cauchy (line 2183)";
+static char __pyx_k_245[] = "\n        standard_cauchy(size=None)\n\n        Standard Cauchy distribution with mode = 0.\n\n        Also known as the Lorentz distribution.\n\n        Parameters\n        ----------\n        size : int or tuple of ints\n            Shape of the output.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The drawn samples.\n\n        Notes\n        -----\n        The probability density function for the full Cauchy distribution is\n\n        .. math:: P(x; x_0, \\gamma) = \\frac{1}{\\pi \\gamma \\bigl[ 1+\n                  (\\frac{x-x_0}{\\gamma})^2 \\bigr] }\n\n        and the Standard Cauchy distribution just sets :math:`x_0=0` and\n        :math:`\\gamma=1`\n\n        The Cauchy distribution arises in the solution to the driven harmonic\n        oscillator problem, and also describes spectral line broadening. It\n        also describes the distribution of values at which a line tilted at\n        a random angle will cut the x axis.\n\n        When studying hypothesis tests that assume normality, seeing how the\n        tests perform on data from a Cauchy distribution is a good indicator of\n        their sensitivity to a heavy-tailed distribution, since the Cauchy looks\n        very much like a Gaussian distribution, but with heavier tails.\n\n        References\n        ----------\n        .. [1] NIST/SEMATECH e-Handbook of Statistical Methods, \"Cauchy\n              Distribution\",\n              http://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm\n        .. [2] Weisstein, Eric W. \"Cauchy Distribution.\" From MathWorld--A\n              Wolfram Web Resource.\n              http://mathworld.wolfram.com/CauchyDistribution.html\n        .. [3] Wikipedia, \"Cauchy distribution\"\n              http://en.wikipedia.org/wiki/Cauchy_distribution\n\n        Examples\n        --------\n        Draw samples and plot the distribution:\n\n        >>> s = np.random.standard_cauchy(1000000)\n        >>> s = s[(s>-25) & (s<""25)]  # truncate distribution so it plots well\n        >>> plt.hist(s, bins=100)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_246[] = "RandomState.standard_t (line 2244)";
+static char __pyx_k_247[] = "\n        standard_t(df, size=None)\n\n        Standard Student's t distribution with df degrees of freedom.\n\n        A special case of the hyperbolic distribution.\n        As `df` gets large, the result resembles that of the standard normal\n        distribution (`standard_normal`).\n\n        Parameters\n        ----------\n        df : int\n            Degrees of freedom, should be > 0.\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            Drawn samples.\n\n        Notes\n        -----\n        The probability density function for the t distribution is\n\n        .. math:: P(x, df) = \\frac{\\Gamma(\\frac{df+1}{2})}{\\sqrt{\\pi df}\n                  \\Gamma(\\frac{df}{2})}\\Bigl( 1+\\frac{x^2}{df} \\Bigr)^{-(df+1)/2}\n\n        The t test is based on an assumption that the data come from a Normal\n        distribution. The t test provides a way to test whether the sample mean\n        (that is the mean calculated from the data) is a good estimate of the true\n        mean.\n\n        The derivation of the t-distribution was forst published in 1908 by William\n        Gisset while working for the Guinness Brewery in Dublin. Due to proprietary\n        issues, he had to publish under a pseudonym, and so he used the name\n        Student.\n\n        References\n        ----------\n        .. [1] Dalgaard, Peter, \"Introductory Statistics With R\",\n               Springer, 2002.\n        .. [2] Wikipedia, \"Student's t-distribution\"\n               http://en.wikipedia.org/wiki/Student's_t-distribution\n\n        Examples\n        --------\n        From Dalgaard page 83 [1]_, suppose the daily energy intake for 11\n        women in Kj is:\n\n        >>> intake = np.array([5260., 5470, 5640, 6180, 6390, 6515, 6805, 7515, \\\n        ...                    7515, 8230, 8770])\n\n        Doe""s their energy intake deviate systematically from the recommended\n        value of 7725 kJ?\n\n        We have 10 degrees of freedom, so is the sample mean within 95% of the\n        recommended value?\n\n        >>> s = np.random.standard_t(10, size=100000)\n        >>> np.mean(intake)\n        6753.636363636364\n        >>> intake.std(ddof=1)\n        1142.1232221373727\n\n        Calculate the t statistic, setting the ddof parameter to the unbiased\n        value so the divisor in the standard deviation will be degrees of\n        freedom, N-1.\n\n        >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake)))\n        >>> import matplotlib.pyplot as plt\n        >>> h = plt.hist(s, bins=100, normed=True)\n\n        For a one-sided t-test, how far out in the distribution does the t\n        statistic appear?\n\n        >>> >>> np.sum(s<t) / float(len(s))\n        0.0090699999999999999  #random\n\n        So the p-value is about 0.009, which says the null hypothesis has a\n        probability of about 99% of being true.\n\n        ";
+static char __pyx_k_248[] = "RandomState.vonmises (line 2345)";
+static char __pyx_k_249[] = "\n        vonmises(mu, kappa, size=None)\n\n        Draw samples from a von Mises distribution.\n\n        Samples are drawn from a von Mises distribution with specified mode\n        (mu) and dispersion (kappa), on the interval [-pi, pi].\n\n        The von Mises distribution (also known as the circular normal\n        distribution) is a continuous probability distribution on the unit\n        circle.  It may be thought of as the circular analogue of the normal\n        distribution.\n\n        Parameters\n        ----------\n        mu : float\n            Mode (\"center\") of the distribution.\n        kappa : float\n            Dispersion of the distribution, has to be >=0.\n        size : int or tuple of int\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : scalar or ndarray\n            The returned samples, which are in the interval [-pi, pi].\n\n        See Also\n        --------\n        scipy.stats.distributions.vonmises : probability density function,\n            distribution, or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the von Mises distribution is\n\n        .. math:: p(x) = \\frac{e^{\\kappa cos(x-\\mu)}}{2\\pi I_0(\\kappa)},\n\n        where :math:`\\mu` is the mode and :math:`\\kappa` the dispersion,\n        and :math:`I_0(\\kappa)` is the modified Bessel function of order 0.\n\n        The von Mises is named for Richard Edler von Mises, who was born in\n        Austria-Hungary, in what is now the Ukraine.  He fled to the United\n        States in 1939 and became a professor at Harvard.  He worked in\n        probability theory, aerodynamics, fluid mechanics, and philosophy of\n        science.\n\n        References\n        ----------\n        Abramowitz, M. and Stegun, I. A. (ed.), *Handbook of Mathematical\n        Functions*, New York: Dover, 1965.\n\n      ""  von Mises, R., *Mathematical Theory of Probability and Statistics*,\n        New York: Academic Press, 1964.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, kappa = 0.0, 4.0 # mean and dispersion\n        >>> s = np.random.vonmises(mu, kappa, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> import scipy.special as sps\n        >>> count, bins, ignored = plt.hist(s, 50, normed=True)\n        >>> x = np.arange(-np.pi, np.pi, 2*np.pi/50.)\n        >>> y = -np.exp(kappa*np.cos(x-mu))/(2*np.pi*sps.jn(0,kappa))\n        >>> plt.plot(x, y/max(y), linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_250[] = "RandomState.pareto (line 2439)";
+static char __pyx_k_251[] = "\n        pareto(a, size=None)\n\n        Draw samples from a Pareto II or Lomax distribution with specified shape.\n\n        The Lomax or Pareto II distribution is a shifted Pareto distribution. The\n        classical Pareto distribution can be obtained from the Lomax distribution\n        by adding the location parameter m, see below. The smallest value of the\n        Lomax distribution is zero while for the classical Pareto distribution it\n        is m, where the standard Pareto distribution has location m=1.\n        Lomax can also be considered as a simplified version of the Generalized\n        Pareto distribution (available in SciPy), with the scale set to one and\n        the location set to zero.\n\n        The Pareto distribution must be greater than zero, and is unbounded above.\n        It is also known as the \"80-20 rule\".  In this distribution, 80 percent of\n        the weights are in the lowest 20 percent of the range, while the other 20\n        percent fill the remaining 80 percent of the range.\n\n        Parameters\n        ----------\n        shape : float, > 0.\n            Shape of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        See Also\n        --------\n        scipy.stats.distributions.lomax.pdf : probability density function,\n            distribution or cumulative density function, etc.\n        scipy.stats.distributions.genpareto.pdf : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Pareto distribution is\n\n        .. math:: p(x) = \\frac{am^a}{x^{a+1}}\n\n        where :math:`a` is the shape and :math:`m` the location\n\n        The Pareto distribution, named after the Italian economist Vilfredo Pareto,\n        is a power law probability distribution useful in many real world probl""ems.\n        Outside the field of economics it is generally referred to as the Bradford\n        distribution. Pareto developed the distribution to describe the\n        distribution of wealth in an economy.  It has also found use in insurance,\n        web page access statistics, oil field sizes, and many other problems,\n        including the download frequency for projects in Sourceforge [1].  It is\n        one of the so-called \"fat-tailed\" distributions.\n\n\n        References\n        ----------\n        .. [1] Francis Hunt and Paul Johnson, On the Pareto Distribution of\n               Sourceforge projects.\n        .. [2] Pareto, V. (1896). Course of Political Economy. Lausanne.\n        .. [3] Reiss, R.D., Thomas, M.(2001), Statistical Analysis of Extreme\n               Values, Birkhauser Verlag, Basel, pp 23-30.\n        .. [4] Wikipedia, \"Pareto distribution\",\n               http://en.wikipedia.org/wiki/Pareto_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a, m = 3., 1. # shape and mode\n        >>> s = np.random.pareto(a, 1000) + m\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 100, normed=True, align='center')\n        >>> fit = a*m**a/bins**(a+1)\n        >>> plt.plot(bins, max(count)*fit/max(fit),linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_252[] = "RandomState.weibull (line 2535)";
+static char __pyx_k_253[] = "\n        weibull(a, size=None)\n\n        Weibull distribution.\n\n        Draw samples from a 1-parameter Weibull distribution with the given\n        shape parameter `a`.\n\n        .. math:: X = (-ln(U))^{1/a}\n\n        Here, U is drawn from the uniform distribution over (0,1].\n\n        The more common 2-parameter Weibull, including a scale parameter\n        :math:`\\lambda` is just :math:`X = \\lambda(-ln(U))^{1/a}`.\n\n        Parameters\n        ----------\n        a : float\n            Shape of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        See Also\n        --------\n        scipy.stats.distributions.weibull_max\n        scipy.stats.distributions.weibull_min\n        scipy.stats.distributions.genextreme\n        gumbel\n\n        Notes\n        -----\n        The Weibull (or Type III asymptotic extreme value distribution for smallest\n        values, SEV Type III, or Rosin-Rammler distribution) is one of a class of\n        Generalized Extreme Value (GEV) distributions used in modeling extreme\n        value problems.  This class includes the Gumbel and Frechet distributions.\n\n        The probability density for the Weibull distribution is\n\n        .. math:: p(x) = \\frac{a}\n                         {\\lambda}(\\frac{x}{\\lambda})^{a-1}e^{-(x/\\lambda)^a},\n\n        where :math:`a` is the shape and :math:`\\lambda` the scale.\n\n        The function has its peak (the mode) at\n        :math:`\\lambda(\\frac{a-1}{a})^{1/a}`.\n\n        When ``a = 1``, the Weibull distribution reduces to the exponential\n        distribution.\n\n        References\n        ----------\n        .. [1] Waloddi Weibull, Professor, Royal Technical University, Stockholm,\n               1939 \"A Statistical Theory Of The Strength Of Materials\",\n               Ingeniorsvetenskapsakademiens Handlingar Nr 151, 1939,\n               General""stabens Litografiska Anstalts Forlag, Stockholm.\n        .. [2] Waloddi Weibull, 1951 \"A Statistical Distribution Function of Wide\n               Applicability\",  Journal Of Applied Mechanics ASME Paper.\n        .. [3] Wikipedia, \"Weibull distribution\",\n               http://en.wikipedia.org/wiki/Weibull_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = 5. # shape\n        >>> s = np.random.weibull(a, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> x = np.arange(1,100.)/50.\n        >>> def weib(x,n,a):\n        ...     return (a / n) * (x / n)**(a - 1) * np.exp(-(x / n)**a)\n\n        >>> count, bins, ignored = plt.hist(np.random.weibull(5.,1000))\n        >>> x = np.arange(1,100.)/50.\n        >>> scale = count.max()/weib(x, 1., 5.).max()\n        >>> plt.plot(x, weib(x, 1., 5.)*scale)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_254[] = "RandomState.power (line 2635)";
+static char __pyx_k_255[] = "\n        power(a, size=None)\n\n        Draws samples in [0, 1] from a power distribution with positive\n        exponent a - 1.\n\n        Also known as the power function distribution.\n\n        Parameters\n        ----------\n        a : float\n            parameter, > 0\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n                    ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n            The returned samples lie in [0, 1].\n\n        Raises\n        ------\n        ValueError\n            If a<1.\n\n        Notes\n        -----\n        The probability density function is\n\n        .. math:: P(x; a) = ax^{a-1}, 0 \\le x \\le 1, a>0.\n\n        The power function distribution is just the inverse of the Pareto\n        distribution. It may also be seen as a special case of the Beta\n        distribution.\n\n        It is used, for example, in modeling the over-reporting of insurance\n        claims.\n\n        References\n        ----------\n        .. [1] Christian Kleiber, Samuel Kotz, \"Statistical size distributions\n               in economics and actuarial sciences\", Wiley, 2003.\n        .. [2] Heckert, N. A. and Filliben, James J. (2003). NIST Handbook 148:\n               Dataplot Reference Manual, Volume 2: Let Subcommands and Library\n               Functions\", National Institute of Standards and Technology Handbook\n               Series, June 2003.\n               http://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = 5. # shape\n        >>> samples = 1000\n        >>> s = np.random.power(a, samples)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, bins=""30)\n        >>> x = np.linspace(0, 1, 100)\n        >>> y = a*x**(a-1.)\n        >>> normed_y = samples*np.diff(bins)[0]*y\n        >>> plt.plot(x, normed_y)\n        >>> plt.show()\n\n        Compare the power function distribution to the inverse of the Pareto.\n\n        >>> from scipy import stats\n        >>> rvs = np.random.power(5, 1000000)\n        >>> rvsp = np.random.pareto(5, 1000000)\n        >>> xx = np.linspace(0,1,100)\n        >>> powpdf = stats.powerlaw.pdf(xx,5)\n\n        >>> plt.figure()\n        >>> plt.hist(rvs, bins=50, normed=True)\n        >>> plt.plot(xx,powpdf,'r-')\n        >>> plt.title('np.random.power(5)')\n\n        >>> plt.figure()\n        >>> plt.hist(1./(1.+rvsp), bins=50, normed=True)\n        >>> plt.plot(xx,powpdf,'r-')\n        >>> plt.title('inverse of 1 + np.random.pareto(5)')\n\n        >>> plt.figure()\n        >>> plt.hist(1./(1.+rvsp), bins=50, normed=True)\n        >>> plt.plot(xx,powpdf,'r-')\n        >>> plt.title('inverse of stats.pareto(5)')\n\n        ";
+static char __pyx_k_256[] = "RandomState.laplace (line 2744)";
+static char __pyx_k_257[] = "\n        laplace(loc=0.0, scale=1.0, size=None)\n\n        Draw samples from the Laplace or double exponential distribution with\n        specified location (or mean) and scale (decay).\n\n        The Laplace distribution is similar to the Gaussian/normal distribution,\n        but is sharper at the peak and has fatter tails. It represents the\n        difference between two independent, identically distributed exponential\n        random variables.\n\n        Parameters\n        ----------\n        loc : float\n            The position, :math:`\\mu`, of the distribution peak.\n        scale : float\n            :math:`\\lambda`, the exponential decay.\n\n        Notes\n        -----\n        It has the probability density function\n\n        .. math:: f(x; \\mu, \\lambda) = \\frac{1}{2\\lambda}\n                                       \\exp\\left(-\\frac{|x - \\mu|}{\\lambda}\\right).\n\n        The first law of Laplace, from 1774, states that the frequency of an error\n        can be expressed as an exponential function of the absolute magnitude of\n        the error, which leads to the Laplace distribution. For many problems in\n        Economics and Health sciences, this distribution seems to model the data\n        better than the standard Gaussian distribution\n\n\n        References\n        ----------\n        .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). Handbook of Mathematical\n               Functions with Formulas, Graphs, and Mathematical Tables, 9th\n               printing.  New York: Dover, 1972.\n\n        .. [2] The Laplace distribution and generalizations\n               By Samuel Kotz, Tomasz J. Kozubowski, Krzysztof Podgorski,\n               Birkhauser, 2001.\n\n        .. [3] Weisstein, Eric W. \"Laplace Distribution.\"\n               From MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/LaplaceDistribution.html\n\n        .. [4] Wikipedia, \"Laplace distribution\",\n               http://en.wikipedia.org/wik""i/Laplace_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution\n\n        >>> loc, scale = 0., 1.\n        >>> s = np.random.laplace(loc, scale, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 30, normed=True)\n        >>> x = np.arange(-8., 8., .01)\n        >>> pdf = np.exp(-abs(x-loc)/scale)/(2.*scale)\n        >>> plt.plot(x, pdf)\n\n        Plot Gaussian for comparison:\n\n        >>> g = (1/(scale * np.sqrt(2 * np.pi)) * \n        ...      np.exp(-(x - loc)**2 / (2 * scale**2)))\n        >>> plt.plot(x,g)\n\n        ";
+static char __pyx_k_258[] = "RandomState.gumbel (line 2834)";
+static char __pyx_k_259[] = "\n        gumbel(loc=0.0, scale=1.0, size=None)\n\n        Gumbel distribution.\n\n        Draw samples from a Gumbel distribution with specified location and scale.\n        For more information on the Gumbel distribution, see Notes and References\n        below.\n\n        Parameters\n        ----------\n        loc : float\n            The location of the mode of the distribution.\n        scale : float\n            The scale parameter of the distribution.\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        out : ndarray\n            The samples\n\n        See Also\n        --------\n        scipy.stats.gumbel_l\n        scipy.stats.gumbel_r\n        scipy.stats.genextreme\n            probability density function, distribution, or cumulative density\n            function, etc. for each of the above\n        weibull\n\n        Notes\n        -----\n        The Gumbel (or Smallest Extreme Value (SEV) or the Smallest Extreme Value\n        Type I) distribution is one of a class of Generalized Extreme Value (GEV)\n        distributions used in modeling extreme value problems.  The Gumbel is a\n        special case of the Extreme Value Type I distribution for maximums from\n        distributions with \"exponential-like\" tails.\n\n        The probability density for the Gumbel distribution is\n\n        .. math:: p(x) = \\frac{e^{-(x - \\mu)/ \\beta}}{\\beta} e^{ -e^{-(x - \\mu)/\n                  \\beta}},\n\n        where :math:`\\mu` is the mode, a location parameter, and :math:`\\beta` is\n        the scale parameter.\n\n        The Gumbel (named for German mathematician Emil Julius Gumbel) was used\n        very early in the hydrology literature, for modeling the occurrence of\n        flood events. It is also used for modeling maximum wind speed and rainfall\n        rates.  It is a \"fat-tailed\" distribution - the ""probability of an event in\n        the tail of the distribution is larger than if one used a Gaussian, hence\n        the surprisingly frequent occurrence of 100-year floods. Floods were\n        initially modeled as a Gaussian process, which underestimated the frequency\n        of extreme events.\n\n\n        It is one of a class of extreme value distributions, the Generalized\n        Extreme Value (GEV) distributions, which also includes the Weibull and\n        Frechet.\n\n        The function has a mean of :math:`\\mu + 0.57721\\beta` and a variance of\n        :math:`\\frac{\\pi^2}{6}\\beta^2`.\n\n        References\n        ----------\n        Gumbel, E. J., *Statistics of Extremes*, New York: Columbia University\n        Press, 1958.\n\n        Reiss, R.-D. and Thomas, M., *Statistical Analysis of Extreme Values from\n        Insurance, Finance, Hydrology and Other Fields*, Basel: Birkhauser Verlag,\n        2001.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, beta = 0, 0.1 # location and scale\n        >>> s = np.random.gumbel(mu, beta, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 30, normed=True)\n        >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta)\n        ...          * np.exp( -np.exp( -(bins - mu) /beta) ),\n        ...          linewidth=2, color='r')\n        >>> plt.show()\n\n        Show how an extreme value distribution can arise from a Gaussian process\n        and compare to a Gaussian:\n\n        >>> means = []\n        >>> maxima = []\n        >>> for i in range(0,1000) :\n        ...    a = np.random.normal(mu, beta, 1000)\n        ...    means.append(a.mean())\n        ...    maxima.append(a.max())\n        >>> count, bins, ignored = plt.hist(maxima, 30, normed=True)\n        >>> beta = np.std(maxima)*np.pi/np.sqrt(6)""\n        >>> mu = np.mean(maxima) - 0.57721*beta\n        >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta)\n        ...          * np.exp(-np.exp(-(bins - mu)/beta)),\n        ...          linewidth=2, color='r')\n        >>> plt.plot(bins, 1/(beta * np.sqrt(2 * np.pi))\n        ...          * np.exp(-(bins - mu)**2 / (2 * beta**2)),\n        ...          linewidth=2, color='g')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_260[] = "RandomState.logistic (line 2965)";
+static char __pyx_k_261[] = "\n        logistic(loc=0.0, scale=1.0, size=None)\n\n        Draw samples from a Logistic distribution.\n\n        Samples are drawn from a Logistic distribution with specified\n        parameters, loc (location or mean, also median), and scale (>0).\n\n        Parameters\n        ----------\n        loc : float\n\n        scale : float > 0.\n\n        size : {tuple, int}\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n                  where the values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.logistic : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Logistic distribution is\n\n        .. math:: P(x) = P(x) = \\frac{e^{-(x-\\mu)/s}}{s(1+e^{-(x-\\mu)/s})^2},\n\n        where :math:`\\mu` = location and :math:`s` = scale.\n\n        The Logistic distribution is used in Extreme Value problems where it\n        can act as a mixture of Gumbel distributions, in Epidemiology, and by\n        the World Chess Federation (FIDE) where it is used in the Elo ranking\n        system, assuming the performance of each player is a logistically\n        distributed random variable.\n\n        References\n        ----------\n        .. [1] Reiss, R.-D. and Thomas M. (2001), Statistical Analysis of Extreme\n               Values, from Insurance, Finance, Hydrology and Other Fields,\n               Birkhauser Verlag, Basel, pp 132-133.\n        .. [2] Weisstein, Eric W. \"Logistic Distribution.\" From\n               MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/LogisticDistribution.html\n        .. [3] Wikipedia, \"Logistic-distribution\",\n               http://en.wikipedia.org/wiki/Logistic-distribution\n\n        Examples\n   ""     --------\n        Draw samples from the distribution:\n\n        >>> loc, scale = 10, 1\n        >>> s = np.random.logistic(loc, scale, 10000)\n        >>> count, bins, ignored = plt.hist(s, bins=50)\n\n        #   plot against distribution\n\n        >>> def logist(x, loc, scale):\n        ...     return exp((loc-x)/scale)/(scale*(1+exp((loc-x)/scale))**2)\n        >>> plt.plot(bins, logist(bins, loc, scale)*count.max()/\\\n        ... logist(bins, loc, scale).max())\n        >>> plt.show()\n\n        ";
+static char __pyx_k_262[] = "RandomState.lognormal (line 3053)";
+static char __pyx_k_263[] = "\n        lognormal(mean=0.0, sigma=1.0, size=None)\n\n        Return samples drawn from a log-normal distribution.\n\n        Draw samples from a log-normal distribution with specified mean,\n        standard deviation, and array shape.  Note that the mean and standard\n        deviation are not the values for the distribution itself, but of the\n        underlying normal distribution it is derived from.\n\n        Parameters\n        ----------\n        mean : float\n            Mean value of the underlying normal distribution\n        sigma : float, > 0.\n            Standard deviation of the underlying normal distribution\n        size : tuple of ints\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : ndarray or float\n            The desired samples. An array of the same shape as `size` if given,\n            if `size` is None a float is returned.\n\n        See Also\n        --------\n        scipy.stats.lognorm : probability density function, distribution,\n            cumulative density function, etc.\n\n        Notes\n        -----\n        A variable `x` has a log-normal distribution if `log(x)` is normally\n        distributed.  The probability density function for the log-normal\n        distribution is:\n\n        .. math:: p(x) = \\frac{1}{\\sigma x \\sqrt{2\\pi}}\n                         e^{(-\\frac{(ln(x)-\\mu)^2}{2\\sigma^2})}\n\n        where :math:`\\mu` is the mean and :math:`\\sigma` is the standard\n        deviation of the normally distributed logarithm of the variable.\n        A log-normal distribution results if a random variable is the *product*\n        of a large number of independent, identically-distributed variables in\n        the same way that a normal distribution results if the variable is the\n        *sum* of a large number of independent, identically-distributed\n        variables.\n\n        Reference""s\n        ----------\n        Limpert, E., Stahel, W. A., and Abbt, M., \"Log-normal Distributions\n        across the Sciences: Keys and Clues,\" *BioScience*, Vol. 51, No. 5,\n        May, 2001.  http://stat.ethz.ch/~stahel/lognormal/bioscience.pdf\n\n        Reiss, R.D. and Thomas, M., *Statistical Analysis of Extreme Values*,\n        Basel: Birkhauser Verlag, 2001, pp. 31-32.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> mu, sigma = 3., 1. # mean and standard deviation\n        >>> s = np.random.lognormal(mu, sigma, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 100, normed=True, align='mid')\n\n        >>> x = np.linspace(min(bins), max(bins), 10000)\n        >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2))\n        ...        / (x * sigma * np.sqrt(2 * np.pi)))\n\n        >>> plt.plot(x, pdf, linewidth=2, color='r')\n        >>> plt.axis('tight')\n        >>> plt.show()\n\n        Demonstrate that taking the products of random samples from a uniform\n        distribution can be fit well by a log-normal probability density function.\n\n        >>> # Generate a thousand samples: each is the product of 100 random\n        >>> # values, drawn from a normal distribution.\n        >>> b = []\n        >>> for i in range(1000):\n        ...    a = 10. + np.random.random(100)\n        ...    b.append(np.product(a))\n\n        >>> b = np.array(b) / np.min(b) # scale values to be positive\n        >>> count, bins, ignored = plt.hist(b, 100, normed=True, align='center')\n        >>> sigma = np.std(np.log(b))\n        >>> mu = np.mean(np.log(b))\n\n        >>> x = np.linspace(min(bins), max(bins), 10000)\n        >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2))\n        ...        / (x * sigma * np.sqrt(2 * np.pi)))\n\n        >>> plt.plot(x, pdf, co""lor='r', linewidth=2)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_264[] = "RandomState.rayleigh (line 3174)";
+static char __pyx_k_265[] = "\n        rayleigh(scale=1.0, size=None)\n\n        Draw samples from a Rayleigh distribution.\n\n        The :math:`\\chi` and Weibull distributions are generalizations of the\n        Rayleigh.\n\n        Parameters\n        ----------\n        scale : scalar\n            Scale, also equals the mode. Should be >= 0.\n        size : int or tuple of ints, optional\n            Shape of the output. Default is None, in which case a single\n            value is returned.\n\n        Notes\n        -----\n        The probability density function for the Rayleigh distribution is\n\n        .. math:: P(x;scale) = \\frac{x}{scale^2}e^{\\frac{-x^2}{2 \\cdotp scale^2}}\n\n        The Rayleigh distribution arises if the wind speed and wind direction are\n        both gaussian variables, then the vector wind velocity forms a Rayleigh\n        distribution. The Rayleigh distribution is used to model the expected\n        output from wind turbines.\n\n        References\n        ----------\n        .. [1] Brighton Webs Ltd., Rayleigh Distribution,\n              http://www.brighton-webs.co.uk/distributions/rayleigh.asp\n        .. [2] Wikipedia, \"Rayleigh distribution\"\n              http://en.wikipedia.org/wiki/Rayleigh_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram\n\n        >>> values = hist(np.random.rayleigh(3, 100000), bins=200, normed=True)\n\n        Wave heights tend to follow a Rayleigh distribution. If the mean wave\n        height is 1 meter, what fraction of waves are likely to be larger than 3\n        meters?\n\n        >>> meanvalue = 1\n        >>> modevalue = np.sqrt(2 / np.pi) * meanvalue\n        >>> s = np.random.rayleigh(modevalue, 1000000)\n\n        The percentage of waves larger than 3 meters is:\n\n        >>> 100.*sum(s>3)/1000000.\n        0.087300000000000003\n\n        ";
+static char __pyx_k_266[] = "RandomState.wald (line 3246)";
+static char __pyx_k_267[] = "\n        wald(mean, scale, size=None)\n\n        Draw samples from a Wald, or Inverse Gaussian, distribution.\n\n        As the scale approaches infinity, the distribution becomes more like a\n        Gaussian.\n\n        Some references claim that the Wald is an Inverse Gaussian with mean=1, but\n        this is by no means universal.\n\n        The Inverse Gaussian distribution was first studied in relationship to\n        Brownian motion. In 1956 M.C.K. Tweedie used the name Inverse Gaussian\n        because there is an inverse relationship between the time to cover a unit\n        distance and distance covered in unit time.\n\n        Parameters\n        ----------\n        mean : scalar\n            Distribution mean, should be > 0.\n        scale : scalar\n            Scale parameter, should be >= 0.\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            Drawn sample, all greater than zero.\n\n        Notes\n        -----\n        The probability density function for the Wald distribution is\n\n        .. math:: P(x;mean,scale) = \\sqrt{\\frac{scale}{2\\pi x^3}}e^\n                                    \\frac{-scale(x-mean)^2}{2\\cdotp mean^2x}\n\n        As noted above the Inverse Gaussian distribution first arise from attempts\n        to model Brownian Motion. It is also a competitor to the Weibull for use in\n        reliability modeling and modeling stock returns and interest rate\n        processes.\n\n        References\n        ----------\n        .. [1] Brighton Webs Ltd., Wald Distribution,\n              http://www.brighton-webs.co.uk/distributions/wald.asp\n        .. [2] Chhikara, Raj S., and Folks, J. Leroy, \"The Inverse Gaussian\n              Distribution: Theory : Methodology, and Applications\", CRC Press,\n              1988.\n        .. [3] Wikipedia, \"Wald distribu""tion\"\n              http://en.wikipedia.org/wiki/Wald_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram:\n\n        >>> import matplotlib.pyplot as plt\n        >>> h = plt.hist(np.random.wald(3, 2, 100000), bins=200, normed=True)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_268[] = "RandomState.triangular (line 3332)";
+static char __pyx_k_269[] = "\n        triangular(left, mode, right, size=None)\n\n        Draw samples from the triangular distribution.\n\n        The triangular distribution is a continuous probability distribution with\n        lower limit left, peak at mode, and upper limit right. Unlike the other\n        distributions, these parameters directly define the shape of the pdf.\n\n        Parameters\n        ----------\n        left : scalar\n            Lower limit.\n        mode : scalar\n            The value where the peak of the distribution occurs.\n            The value should fulfill the condition ``left <= mode <= right``.\n        right : scalar\n            Upper limit, should be larger than `left`.\n        size : int or tuple of ints, optional\n            Output shape. Default is None, in which case a single value is\n            returned.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The returned samples all lie in the interval [left, right].\n\n        Notes\n        -----\n        The probability density function for the Triangular distribution is\n\n        .. math:: P(x;l, m, r) = \\begin{cases}\n                  \\frac{2(x-l)}{(r-l)(m-l)}& \\text{for $l \\leq x \\leq m$},\\\\\n                  \\frac{2(m-x)}{(r-l)(r-m)}& \\text{for $m \\leq x \\leq r$},\\\\\n                  0& \\text{otherwise}.\n                  \\end{cases}\n\n        The triangular distribution is often used in ill-defined problems where the\n        underlying distribution is not known, but some knowledge of the limits and\n        mode exists. Often it is used in simulations.\n\n        References\n        ----------\n        .. [1] Wikipedia, \"Triangular distribution\"\n              http://en.wikipedia.org/wiki/Triangular_distribution\n\n        Examples\n        --------\n        Draw values from the distribution and plot the histogram:\n\n        >>> import matplotlib.pyplot as plt\n        >>> h = plt.hist(np.random.triangular(-3, 0, 8, 100000), bins=""200,\n        ...              normed=True)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_270[] = "RandomState.binomial (line 3420)";
+static char __pyx_k_271[] = "\n        binomial(n, p, size=None)\n\n        Draw samples from a binomial distribution.\n\n        Samples are drawn from a Binomial distribution with specified\n        parameters, n trials and p probability of success where\n        n an integer >= 0 and p is in the interval [0,1]. (n may be\n        input as a float, but it is truncated to an integer in use)\n\n        Parameters\n        ----------\n        n : float (but truncated to an integer)\n                parameter, >= 0.\n        p : float\n                parameter, >= 0 and <=1.\n        size : {tuple, int}\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n                  where the values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.binom : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Binomial distribution is\n\n        .. math:: P(N) = \\binom{n}{N}p^N(1-p)^{n-N},\n\n        where :math:`n` is the number of trials, :math:`p` is the probability\n        of success, and :math:`N` is the number of successes.\n\n        When estimating the standard error of a proportion in a population by\n        using a random sample, the normal distribution works well unless the\n        product p*n <=5, where p = population proportion estimate, and n =\n        number of samples, in which case the binomial distribution is used\n        instead. For example, a sample of 15 people shows 4 who are left\n        handed, and 11 who are right handed. Then p = 4/15 = 27%. 0.27*15 = 4,\n        so the binomial distribution should be used in this case.\n\n        References\n        ----------\n        .. [1] Dalgaard, Peter, \"Introductory Statistics with R\",\n               Springer-Verlag, 2002.""\n        .. [2] Glantz, Stanton A. \"Primer of Biostatistics.\", McGraw-Hill,\n               Fifth Edition, 2002.\n        .. [3] Lentner, Marvin, \"Elementary Applied Statistics\", Bogden\n               and Quigley, 1972.\n        .. [4] Weisstein, Eric W. \"Binomial Distribution.\" From MathWorld--A\n               Wolfram Web Resource.\n               http://mathworld.wolfram.com/BinomialDistribution.html\n        .. [5] Wikipedia, \"Binomial-distribution\",\n               http://en.wikipedia.org/wiki/Binomial_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> n, p = 10, .5 # number of trials, probability of each trial\n        >>> s = np.random.binomial(n, p, 1000)\n        # result of flipping a coin 10 times, tested 1000 times.\n\n        A real world example. A company drills 9 wild-cat oil exploration\n        wells, each with an estimated probability of success of 0.1. All nine\n        wells fail. What is the probability of that happening?\n\n        Let's do 20,000 trials of the model, and count the number that\n        generate zero positive results.\n\n        >>> sum(np.random.binomial(9,0.1,20000)==0)/20000.\n        answer = 0.38885, or 38%.\n\n        ";
+static char __pyx_k_272[] = "RandomState.negative_binomial (line 3530)";
+static char __pyx_k_273[] = "\n        negative_binomial(n, p, size=None)\n\n        Draw samples from a negative_binomial distribution.\n\n        Samples are drawn from a negative_Binomial distribution with specified\n        parameters, `n` trials and `p` probability of success where `n` is an\n        integer > 0 and `p` is in the interval [0, 1].\n\n        Parameters\n        ----------\n        n : int\n            Parameter, > 0.\n        p : float\n            Parameter, >= 0 and <=1.\n        size : int or tuple of ints\n            Output shape. If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : int or ndarray of ints\n            Drawn samples.\n\n        Notes\n        -----\n        The probability density for the Negative Binomial distribution is\n\n        .. math:: P(N;n,p) = \\binom{N+n-1}{n-1}p^{n}(1-p)^{N},\n\n        where :math:`n-1` is the number of successes, :math:`p` is the probability\n        of success, and :math:`N+n-1` is the number of trials.\n\n        The negative binomial distribution gives the probability of n-1 successes\n        and N failures in N+n-1 trials, and success on the (N+n)th trial.\n\n        If one throws a die repeatedly until the third time a \"1\" appears, then the\n        probability distribution of the number of non-\"1\"s that appear before the\n        third \"1\" is a negative binomial distribution.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Negative Binomial Distribution.\" From\n               MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/NegativeBinomialDistribution.html\n        .. [2] Wikipedia, \"Negative binomial distribution\",\n               http://en.wikipedia.org/wiki/Negative_binomial_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        A real world example. A company drills wild-cat oil exploration well""s, each\n        with an estimated probability of success of 0.1.  What is the probability\n        of having one success for each successive well, that is what is the\n        probability of a single success after drilling 5 wells, after 6 wells,\n        etc.?\n\n        >>> s = np.random.negative_binomial(1, 0.1, 100000)\n        >>> for i in range(1, 11):\n        ...    probability = sum(s<i) / 100000.\n        ...    print i, \"wells drilled, probability of one success =\", probability\n\n        ";
+static char __pyx_k_274[] = "RandomState.poisson (line 3625)";
+static char __pyx_k_275[] = "\n        poisson(lam=1.0, size=None)\n\n        Draw samples from a Poisson distribution.\n\n        The Poisson distribution is the limit of the Binomial\n        distribution for large N.\n\n        Parameters\n        ----------\n        lam : float\n            Expectation of interval, should be >= 0.\n        size : int or tuple of ints, optional\n            Output shape. If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Notes\n        -----\n        The Poisson distribution\n\n        .. math:: f(k; \\lambda)=\\frac{\\lambda^k e^{-\\lambda}}{k!}\n\n        For events with an expected separation :math:`\\lambda` the Poisson\n        distribution :math:`f(k; \\lambda)` describes the probability of\n        :math:`k` events occurring within the observed interval :math:`\\lambda`.\n\n        Because the output is limited to the range of the C long type, a\n        ValueError is raised when `lam` is within 10 sigma of the maximum\n        representable value.\n\n        References\n        ----------\n        .. [1] Weisstein, Eric W. \"Poisson Distribution.\" From MathWorld--A Wolfram\n               Web Resource. http://mathworld.wolfram.com/PoissonDistribution.html\n        .. [2] Wikipedia, \"Poisson distribution\",\n           http://en.wikipedia.org/wiki/Poisson_distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> import numpy as np\n        >>> s = np.random.poisson(5, 10000)\n\n        Display histogram of the sample:\n\n        >>> import matplotlib.pyplot as plt\n        >>> count, bins, ignored = plt.hist(s, 14, normed=True)\n        >>> plt.show()\n\n        ";
+static char __pyx_k_276[] = "RandomState.zipf (line 3696)";
+static char __pyx_k_277[] = "\n        zipf(a, size=None)\n\n        Draw samples from a Zipf distribution.\n\n        Samples are drawn from a Zipf distribution with specified parameter\n        `a` > 1.\n\n        The Zipf distribution (also known as the zeta distribution) is a\n        continuous probability distribution that satisfies Zipf's law: the\n        frequency of an item is inversely proportional to its rank in a\n        frequency table.\n\n        Parameters\n        ----------\n        a : float > 1\n            Distribution parameter.\n        size : int or tuple of int, optional\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn; a single integer is equivalent in\n            its result to providing a mono-tuple, i.e., a 1-D array of length\n            *size* is returned.  The default is None, in which case a single\n            scalar is returned.\n\n        Returns\n        -------\n        samples : scalar or ndarray\n            The returned samples are greater than or equal to one.\n\n        See Also\n        --------\n        scipy.stats.distributions.zipf : probability density function,\n            distribution, or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Zipf distribution is\n\n        .. math:: p(x) = \\frac{x^{-a}}{\\zeta(a)},\n\n        where :math:`\\zeta` is the Riemann Zeta function.\n\n        It is named for the American linguist George Kingsley Zipf, who noted\n        that the frequency of any word in a sample of a language is inversely\n        proportional to its rank in the frequency table.\n\n        References\n        ----------\n        Zipf, G. K., *Selected Studies of the Principle of Relative Frequency\n        in Language*, Cambridge, MA: Harvard Univ. Press, 1932.\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = 2. # parameter\n        >>> s = np.random.zipf""(a, 1000)\n\n        Display the histogram of the samples, along with\n        the probability density function:\n\n        >>> import matplotlib.pyplot as plt\n        >>> import scipy.special as sps\n        Truncate s values at 50 so plot is interesting\n        >>> count, bins, ignored = plt.hist(s[s<50], 50, normed=True)\n        >>> x = np.arange(1., 50.)\n        >>> y = x**(-a)/sps.zetac(a)\n        >>> plt.plot(x, y/max(y), linewidth=2, color='r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_278[] = "RandomState.geometric (line 3784)";
+static char __pyx_k_279[] = "\n        geometric(p, size=None)\n\n        Draw samples from the geometric distribution.\n\n        Bernoulli trials are experiments with one of two outcomes:\n        success or failure (an example of such an experiment is flipping\n        a coin).  The geometric distribution models the number of trials\n        that must be run in order to achieve success.  It is therefore\n        supported on the positive integers, ``k = 1, 2, ...``.\n\n        The probability mass function of the geometric distribution is\n\n        .. math:: f(k) = (1 - p)^{k - 1} p\n\n        where `p` is the probability of success of an individual trial.\n\n        Parameters\n        ----------\n        p : float\n            The probability of success of an individual trial.\n        size : tuple of ints\n            Number of values to draw from the distribution.  The output\n            is shaped according to `size`.\n\n        Returns\n        -------\n        out : ndarray\n            Samples from the geometric distribution, shaped according to\n            `size`.\n\n        Examples\n        --------\n        Draw ten thousand values from the geometric distribution,\n        with the probability of an individual success equal to 0.35:\n\n        >>> z = np.random.geometric(p=0.35, size=10000)\n\n        How many trials succeeded after a single run?\n\n        >>> (z == 1).sum() / 10000.\n        0.34889999999999999 #random\n\n        ";
+static char __pyx_k_280[] = "RandomState.hypergeometric (line 3850)";
+static char __pyx_k_281[] = "\n        hypergeometric(ngood, nbad, nsample, size=None)\n\n        Draw samples from a Hypergeometric distribution.\n\n        Samples are drawn from a Hypergeometric distribution with specified\n        parameters, ngood (ways to make a good selection), nbad (ways to make\n        a bad selection), and nsample = number of items sampled, which is less\n        than or equal to the sum ngood + nbad.\n\n        Parameters\n        ----------\n        ngood : int or array_like\n            Number of ways to make a good selection.  Must be nonnegative.\n        nbad : int or array_like\n            Number of ways to make a bad selection.  Must be nonnegative.\n        nsample : int or array_like\n            Number of items sampled.  Must be at least 1 and at most\n            ``ngood + nbad``.\n        size : int or tuple of int\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : ndarray or scalar\n            The values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.hypergeom : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Hypergeometric distribution is\n\n        .. math:: P(x) = \\frac{\\binom{m}{n}\\binom{N-m}{n-x}}{\\binom{N}{n}},\n\n        where :math:`0 \\le x \\le m` and :math:`n+m-N \\le x \\le n`\n\n        for P(x) the probability of x successes, n = ngood, m = nbad, and\n        N = number of samples.\n\n        Consider an urn with black and white marbles in it, ngood of them\n        black and nbad are white. If you draw nsample balls without\n        replacement, then the Hypergeometric distribution describes the\n        distribution of black balls in the drawn sample.\n\n        Note that this distribution is very similar to the Binomial\n        distrib""ution, except that in this case, samples are drawn without\n        replacement, whereas in the Binomial case samples are drawn with\n        replacement (or the sample space is infinite). As the sample space\n        becomes large, this distribution approaches the Binomial.\n\n        References\n        ----------\n        .. [1] Lentner, Marvin, \"Elementary Applied Statistics\", Bogden\n               and Quigley, 1972.\n        .. [2] Weisstein, Eric W. \"Hypergeometric Distribution.\" From\n               MathWorld--A Wolfram Web Resource.\n               http://mathworld.wolfram.com/HypergeometricDistribution.html\n        .. [3] Wikipedia, \"Hypergeometric-distribution\",\n               http://en.wikipedia.org/wiki/Hypergeometric-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> ngood, nbad, nsamp = 100, 2, 10\n        # number of good, number of bad, and number of samples\n        >>> s = np.random.hypergeometric(ngood, nbad, nsamp, 1000)\n        >>> hist(s)\n        #   note that it is very unlikely to grab both bad items\n\n        Suppose you have an urn with 15 white and 15 black marbles.\n        If you pull 15 marbles at random, how likely is it that\n        12 or more of them are one color?\n\n        >>> s = np.random.hypergeometric(15, 15, 15, 100000)\n        >>> sum(s>=12)/100000. + sum(s<=3)/100000.\n        #   answer = 0.003 ... pretty unlikely!\n\n        ";
+static char __pyx_k_282[] = "RandomState.logseries (line 3969)";
+static char __pyx_k_283[] = "\n        logseries(p, size=None)\n\n        Draw samples from a Logarithmic Series distribution.\n\n        Samples are drawn from a Log Series distribution with specified\n        parameter, p (probability, 0 < p < 1).\n\n        Parameters\n        ----------\n        loc : float\n\n        scale : float > 0.\n\n        size : {tuple, int}\n            Output shape.  If the given shape is, e.g., ``(m, n, k)``, then\n            ``m * n * k`` samples are drawn.\n\n        Returns\n        -------\n        samples : {ndarray, scalar}\n                  where the values are all integers in  [0, n].\n\n        See Also\n        --------\n        scipy.stats.distributions.logser : probability density function,\n            distribution or cumulative density function, etc.\n\n        Notes\n        -----\n        The probability density for the Log Series distribution is\n\n        .. math:: P(k) = \\frac{-p^k}{k \\ln(1-p)},\n\n        where p = probability.\n\n        The Log Series distribution is frequently used to represent species\n        richness and occurrence, first proposed by Fisher, Corbet, and\n        Williams in 1943 [2].  It may also be used to model the numbers of\n        occupants seen in cars [3].\n\n        References\n        ----------\n        .. [1] Buzas, Martin A.; Culver, Stephen J.,  Understanding regional\n               species diversity through the log series distribution of\n               occurrences: BIODIVERSITY RESEARCH Diversity & Distributions,\n               Volume 5, Number 5, September 1999 , pp. 187-195(9).\n        .. [2] Fisher, R.A,, A.S. Corbet, and C.B. Williams. 1943. The\n               relation between the number of species and the number of\n               individuals in a random sample of an animal population.\n               Journal of Animal Ecology, 12:42-58.\n        .. [3] D. J. Hand, F. Daly, D. Lunn, E. Ostrowski, A Handbook of Small\n               Data Sets, CRC Press, 1994.\n        .. [4] Wikipedia, \"Log""arithmic-distribution\",\n               http://en.wikipedia.org/wiki/Logarithmic-distribution\n\n        Examples\n        --------\n        Draw samples from the distribution:\n\n        >>> a = .6\n        >>> s = np.random.logseries(a, 10000)\n        >>> count, bins, ignored = plt.hist(s)\n\n        #   plot against distribution\n\n        >>> def logseries(k, p):\n        ...     return -p**k/(k*log(1-p))\n        >>> plt.plot(bins, logseries(bins, a)*count.max()/\n                     logseries(bins, a).max(), 'r')\n        >>> plt.show()\n\n        ";
+static char __pyx_k_284[] = "RandomState.multivariate_normal (line 4064)";
+static char __pyx_k_285[] = "\n        multivariate_normal(mean, cov[, size])\n\n        Draw random samples from a multivariate normal distribution.\n\n        The multivariate normal, multinormal or Gaussian distribution is a\n        generalization of the one-dimensional normal distribution to higher\n        dimensions.  Such a distribution is specified by its mean and\n        covariance matrix.  These parameters are analogous to the mean\n        (average or \"center\") and variance (standard deviation, or \"width,\"\n        squared) of the one-dimensional normal distribution.\n\n        Parameters\n        ----------\n        mean : 1-D array_like, of length N\n            Mean of the N-dimensional distribution.\n        cov : 2-D array_like, of shape (N, N)\n            Covariance matrix of the distribution.  Must be symmetric and\n            positive-semidefinite for \"physically meaningful\" results.\n        size : int or tuple of ints, optional\n            Given a shape of, for example, ``(m,n,k)``, ``m*n*k`` samples are\n            generated, and packed in an `m`-by-`n`-by-`k` arrangement.  Because\n            each sample is `N`-dimensional, the output shape is ``(m,n,k,N)``.\n            If no shape is specified, a single (`N`-D) sample is returned.\n\n        Returns\n        -------\n        out : ndarray\n            The drawn samples, of shape *size*, if that was provided.  If not,\n            the shape is ``(N,)``.\n\n            In other words, each entry ``out[i,j,...,:]`` is an N-dimensional\n            value drawn from the distribution.\n\n        Notes\n        -----\n        The mean is a coordinate in N-dimensional space, which represents the\n        location where samples are most likely to be generated.  This is\n        analogous to the peak of the bell curve for the one-dimensional or\n        univariate normal distribution.\n\n        Covariance indicates the level to which two variables vary together.\n        From the multivariate normal distribution, we"" draw N-dimensional\n        samples, :math:`X = [x_1, x_2, ... x_N]`.  The covariance matrix\n        element :math:`C_{ij}` is the covariance of :math:`x_i` and :math:`x_j`.\n        The element :math:`C_{ii}` is the variance of :math:`x_i` (i.e. its\n        \"spread\").\n\n        Instead of specifying the full covariance matrix, popular\n        approximations include:\n\n          - Spherical covariance (*cov* is a multiple of the identity matrix)\n          - Diagonal covariance (*cov* has non-negative elements, and only on\n            the diagonal)\n\n        This geometrical property can be seen in two dimensions by plotting\n        generated data-points:\n\n        >>> mean = [0,0]\n        >>> cov = [[1,0],[0,100]] # diagonal covariance, points lie on x or y-axis\n\n        >>> import matplotlib.pyplot as plt\n        >>> x,y = np.random.multivariate_normal(mean,cov,5000).T\n        >>> plt.plot(x,y,'x'); plt.axis('equal'); plt.show()\n\n        Note that the covariance matrix must be non-negative definite.\n\n        References\n        ----------\n        Papoulis, A., *Probability, Random Variables, and Stochastic Processes*,\n        3rd ed., New York: McGraw-Hill, 1991.\n\n        Duda, R. O., Hart, P. E., and Stork, D. G., *Pattern Classification*,\n        2nd ed., New York: Wiley, 2001.\n\n        Examples\n        --------\n        >>> mean = (1,2)\n        >>> cov = [[1,0],[1,0]]\n        >>> x = np.random.multivariate_normal(mean,cov,(3,3))\n        >>> x.shape\n        (3, 3, 2)\n\n        The following is probably true, given that 0.6 is roughly twice the\n        standard deviation:\n\n        >>> print list( (x[0,0,:] - mean) < 0.6 )\n        [True, True]\n\n        ";
+static char __pyx_k_286[] = "RandomState.multinomial (line 4207)";
+static char __pyx_k_287[] = "\n        multinomial(n, pvals, size=None)\n\n        Draw samples from a multinomial distribution.\n\n        The multinomial distribution is a multivariate generalisation of the\n        binomial distribution.  Take an experiment with one of ``p``\n        possible outcomes.  An example of such an experiment is throwing a dice,\n        where the outcome can be 1 through 6.  Each sample drawn from the\n        distribution represents `n` such experiments.  Its values,\n        ``X_i = [X_0, X_1, ..., X_p]``, represent the number of times the outcome\n        was ``i``.\n\n        Parameters\n        ----------\n        n : int\n            Number of experiments.\n        pvals : sequence of floats, length p\n            Probabilities of each of the ``p`` different outcomes.  These\n            should sum to 1 (however, the last element is always assumed to\n            account for the remaining probability, as long as\n            ``sum(pvals[:-1]) <= 1)``.\n        size : tuple of ints\n            Given a `size` of ``(M, N, K)``, then ``M*N*K`` samples are drawn,\n            and the output shape becomes ``(M, N, K, p)``, since each sample\n            has shape ``(p,)``.\n\n        Examples\n        --------\n        Throw a dice 20 times:\n\n        >>> np.random.multinomial(20, [1/6.]*6, size=1)\n        array([[4, 1, 7, 5, 2, 1]])\n\n        It landed 4 times on 1, once on 2, etc.\n\n        Now, throw the dice 20 times, and 20 times again:\n\n        >>> np.random.multinomial(20, [1/6.]*6, size=2)\n        array([[3, 4, 3, 3, 4, 3],\n               [2, 4, 3, 4, 0, 7]])\n\n        For the first run, we threw 3 times 1, 4 times 2, etc.  For the second,\n        we threw 2 times 1, 4 times 2, etc.\n\n        A loaded dice is more likely to land on number 6:\n\n        >>> np.random.multinomial(100, [1/7.]*5)\n        array([13, 16, 13, 16, 42])\n\n        ";
+static char __pyx_k_288[] = "RandomState.dirichlet (line 4295)";
+static char __pyx_k_289[] = "\n        dirichlet(alpha, size=None)\n\n        Draw samples from the Dirichlet distribution.\n\n        Draw `size` samples of dimension k from a Dirichlet distribution. A\n        Dirichlet-distributed random variable can be seen as a multivariate\n        generalization of a Beta distribution. Dirichlet pdf is the conjugate\n        prior of a multinomial in Bayesian inference.\n\n        Parameters\n        ----------\n        alpha : array\n            Parameter of the distribution (k dimension for sample of\n            dimension k).\n        size : array\n            Number of samples to draw.\n\n        Returns\n        -------\n        samples : ndarray,\n            The drawn samples, of shape (alpha.ndim, size).\n\n        Notes\n        -----\n        .. math:: X \\approx \\prod_{i=1}^{k}{x^{\\alpha_i-1}_i}\n\n        Uses the following property for computation: for each dimension,\n        draw a random sample y_i from a standard gamma generator of shape\n        `alpha_i`, then\n        :math:`X = \\frac{1}{\\sum_{i=1}^k{y_i}} (y_1, \\ldots, y_n)` is\n        Dirichlet distributed.\n\n        References\n        ----------\n        .. [1] David McKay, \"Information Theory, Inference and Learning\n               Algorithms,\" chapter 23,\n               http://www.inference.phy.cam.ac.uk/mackay/\n        .. [2] Wikipedia, \"Dirichlet distribution\",\n               http://en.wikipedia.org/wiki/Dirichlet_distribution\n\n        Examples\n        --------\n        Taking an example cited in Wikipedia, this distribution can be used if\n        one wanted to cut strings (each of initial length 1.0) into K pieces\n        with different lengths, where each piece had, on average, a designated\n        average length, but allowing some variation in the relative sizes of the\n        pieces.\n\n        >>> s = np.random.dirichlet((10, 5, 3), 20).transpose()\n\n        >>> plt.barh(range(20), s[0])\n        >>> plt.barh(range(20), s[1], left=s[0], color='g')""\n        >>> plt.barh(range(20), s[2], left=s[0]+s[1], color='r')\n        >>> plt.title(\"Lengths of Strings\")\n\n        ";
+static char __pyx_k_290[] = "RandomState.shuffle (line 4407)";
+static char __pyx_k_291[] = "\n        shuffle(x)\n\n        Modify a sequence in-place by shuffling its contents.\n\n        Parameters\n        ----------\n        x : array_like\n            The array or list to be shuffled.\n\n        Returns\n        -------\n        None\n\n        Examples\n        --------\n        >>> arr = np.arange(10)\n        >>> np.random.shuffle(arr)\n        >>> arr\n        [1 7 5 2 9 4 3 6 0 8]\n\n        This function only shuffles the array along the first index of a\n        multi-dimensional array:\n\n        >>> arr = np.arange(9).reshape((3, 3))\n        >>> np.random.shuffle(arr)\n        >>> arr\n        array([[3, 4, 5],\n               [6, 7, 8],\n               [0, 1, 2]])\n\n        ";
+static char __pyx_k_292[] = "RandomState.permutation (line 4466)";
+static char __pyx_k_293[] = "\n        permutation(x)\n\n        Randomly permute a sequence, or return a permuted range.\n\n        If `x` is a multi-dimensional array, it is only shuffled along its\n        first index.\n\n        Parameters\n        ----------\n        x : int or array_like\n            If `x` is an integer, randomly permute ``np.arange(x)``.\n            If `x` is an array, make a copy and shuffle the elements\n            randomly.\n\n        Returns\n        -------\n        out : ndarray\n            Permuted sequence or array range.\n\n        Examples\n        --------\n        >>> np.random.permutation(10)\n        array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6])\n\n        >>> np.random.permutation([1, 4, 9, 12, 15])\n        array([15,  1,  9,  4, 12])\n\n        >>> arr = np.arange(9).reshape((3, 3))\n        >>> np.random.permutation(arr)\n        array([[6, 7, 8],\n               [0, 1, 2],\n               [3, 4, 5]])\n\n        ";
 static char __pyx_k__df[] = "df";
 static char __pyx_k__mu[] = "mu";
 static char __pyx_k__np[] = "np";
@@ -1142,6 +1135,7 @@ static char __pyx_k__equal[] = "equal";
 static char __pyx_k__gamma[] = "gamma";
 static char __pyx_k__iinfo[] = "iinfo";
 static char __pyx_k__index[] = "index";
+static char __pyx_k__isnan[] = "isnan";
 static char __pyx_k__kappa[] = "kappa";
 static char __pyx_k__ngood[] = "ngood";
 static char __pyx_k__numpy[] = "numpy";
@@ -1238,33 +1232,32 @@ static PyObject *__pyx_kp_s_130;
 static PyObject *__pyx_kp_s_135;
 static PyObject *__pyx_kp_s_137;
 static PyObject *__pyx_kp_s_139;
-static PyObject *__pyx_kp_s_144;
+static PyObject *__pyx_kp_s_141;
+static PyObject *__pyx_kp_s_146;
 static PyObject *__pyx_kp_s_15;
-static PyObject *__pyx_kp_s_152;
 static PyObject *__pyx_kp_s_154;
-static PyObject *__pyx_kp_s_157;
+static PyObject *__pyx_kp_s_156;
 static PyObject *__pyx_kp_s_159;
-static PyObject *__pyx_kp_s_162;
+static PyObject *__pyx_kp_s_161;
 static PyObject *__pyx_kp_s_164;
-static PyObject *__pyx_kp_s_168;
+static PyObject *__pyx_kp_s_166;
 static PyObject *__pyx_kp_s_170;
 static PyObject *__pyx_kp_s_172;
 static PyObject *__pyx_kp_s_174;
+static PyObject *__pyx_kp_s_176;
 static PyObject *__pyx_kp_s_18;
-static PyObject *__pyx_kp_s_180;
 static PyObject *__pyx_kp_s_182;
-static PyObject *__pyx_n_s_186;
-static PyObject *__pyx_kp_s_187;
+static PyObject *__pyx_kp_s_184;
+static PyObject *__pyx_n_s_188;
 static PyObject *__pyx_kp_s_189;
 static PyObject *__pyx_kp_s_191;
-static PyObject *__pyx_kp_s_194;
-static PyObject *__pyx_kp_s_197;
+static PyObject *__pyx_kp_s_193;
+static PyObject *__pyx_kp_s_196;
+static PyObject *__pyx_kp_s_199;
 static PyObject *__pyx_kp_s_20;
-static PyObject *__pyx_kp_s_201;
-static PyObject *__pyx_n_s_204;
-static PyObject *__pyx_n_s_205;
-static PyObject *__pyx_kp_u_206;
-static PyObject *__pyx_kp_u_207;
+static PyObject *__pyx_kp_s_203;
+static PyObject *__pyx_n_s_206;
+static PyObject *__pyx_n_s_207;
 static PyObject *__pyx_kp_u_208;
 static PyObject *__pyx_kp_u_209;
 static PyObject *__pyx_kp_u_210;
@@ -1353,6 +1346,8 @@ static PyObject *__pyx_kp_u_288;
 static PyObject *__pyx_kp_u_289;
 static PyObject *__pyx_kp_u_290;
 static PyObject *__pyx_kp_u_291;
+static PyObject *__pyx_kp_u_292;
+static PyObject *__pyx_kp_u_293;
 static PyObject *__pyx_kp_s_30;
 static PyObject *__pyx_kp_s_32;
 static PyObject *__pyx_kp_s_34;
@@ -1426,6 +1421,7 @@ static PyObject *__pyx_n_s__index;
 static PyObject *__pyx_n_s__int;
 static PyObject *__pyx_n_s__integer;
 static PyObject *__pyx_n_s__intp;
+static PyObject *__pyx_n_s__isnan;
 static PyObject *__pyx_n_s__item;
 static PyObject *__pyx_n_s__kappa;
 static PyObject *__pyx_n_s__l;
@@ -1535,7 +1531,7 @@ static PyObject *__pyx_k_107;
 static PyObject *__pyx_k_110;
 static PyObject *__pyx_k_111;
 static PyObject *__pyx_k_116;
-static PyObject *__pyx_k_151;
+static PyObject *__pyx_k_153;
 static PyObject *__pyx_k_tuple_2;
 static PyObject *__pyx_k_tuple_3;
 static PyObject *__pyx_k_tuple_4;
@@ -1600,8 +1596,8 @@ static PyObject *__pyx_k_tuple_94;
 static PyObject *__pyx_k_tuple_95;
 static PyObject *__pyx_k_tuple_96;
 static PyObject *__pyx_k_tuple_97;
-static PyObject *__pyx_k_slice_193;
 static PyObject *__pyx_k_slice_195;
+static PyObject *__pyx_k_slice_197;
 static PyObject *__pyx_k_tuple_100;
 static PyObject *__pyx_k_tuple_101;
 static PyObject *__pyx_k_tuple_104;
@@ -1625,46 +1621,47 @@ static PyObject *__pyx_k_tuple_134;
 static PyObject *__pyx_k_tuple_136;
 static PyObject *__pyx_k_tuple_138;
 static PyObject *__pyx_k_tuple_140;
-static PyObject *__pyx_k_tuple_141;
 static PyObject *__pyx_k_tuple_142;
 static PyObject *__pyx_k_tuple_143;
+static PyObject *__pyx_k_tuple_144;
 static PyObject *__pyx_k_tuple_145;
-static PyObject *__pyx_k_tuple_146;
 static PyObject *__pyx_k_tuple_147;
 static PyObject *__pyx_k_tuple_148;
 static PyObject *__pyx_k_tuple_149;
 static PyObject *__pyx_k_tuple_150;
-static PyObject *__pyx_k_tuple_153;
+static PyObject *__pyx_k_tuple_151;
+static PyObject *__pyx_k_tuple_152;
 static PyObject *__pyx_k_tuple_155;
-static PyObject *__pyx_k_tuple_156;
+static PyObject *__pyx_k_tuple_157;
 static PyObject *__pyx_k_tuple_158;
 static PyObject *__pyx_k_tuple_160;
-static PyObject *__pyx_k_tuple_161;
+static PyObject *__pyx_k_tuple_162;
 static PyObject *__pyx_k_tuple_163;
 static PyObject *__pyx_k_tuple_165;
-static PyObject *__pyx_k_tuple_166;
 static PyObject *__pyx_k_tuple_167;
+static PyObject *__pyx_k_tuple_168;
 static PyObject *__pyx_k_tuple_169;
 static PyObject *__pyx_k_tuple_171;
 static PyObject *__pyx_k_tuple_173;
 static PyObject *__pyx_k_tuple_175;
-static PyObject *__pyx_k_tuple_176;
 static PyObject *__pyx_k_tuple_177;
 static PyObject *__pyx_k_tuple_178;
 static PyObject *__pyx_k_tuple_179;
+static PyObject *__pyx_k_tuple_180;
 static PyObject *__pyx_k_tuple_181;
 static PyObject *__pyx_k_tuple_183;
-static PyObject *__pyx_k_tuple_184;
 static PyObject *__pyx_k_tuple_185;
-static PyObject *__pyx_k_tuple_188;
+static PyObject *__pyx_k_tuple_186;
+static PyObject *__pyx_k_tuple_187;
 static PyObject *__pyx_k_tuple_190;
 static PyObject *__pyx_k_tuple_192;
-static PyObject *__pyx_k_tuple_196;
+static PyObject *__pyx_k_tuple_194;
 static PyObject *__pyx_k_tuple_198;
-static PyObject *__pyx_k_tuple_199;
-static PyObject *__pyx_k_tuple_202;
-static PyObject *__pyx_k_tuple_203;
-static PyObject *__pyx_k_codeobj_200;
+static PyObject *__pyx_k_tuple_200;
+static PyObject *__pyx_k_tuple_201;
+static PyObject *__pyx_k_tuple_204;
+static PyObject *__pyx_k_tuple_205;
+static PyObject *__pyx_k_codeobj_202;
 
 /* "mtrand.pyx":130
  * import warnings
@@ -5400,7 +5397,8 @@ static PyObject *__pyx_pf_6mtrand__shape_from_size(CYTHON_UNUSED PyObject *__pyx
         __Pyx_GOTREF(((PyObject *)__pyx_t_12));
         __Pyx_DECREF(__pyx_t_11); __pyx_t_11 = 0;
         __Pyx_DECREF(((PyObject *)__pyx_t_10)); __pyx_t_10 = 0;
-        __Pyx_XDECREF_SET(__pyx_v_shape, ((PyObject*)__pyx_t_12));
+        __Pyx_XDECREF(((PyObject *)__pyx_v_shape));
+        __pyx_v_shape = ((PyObject*)__pyx_t_12);
         __pyx_t_12 = 0;
         __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
         __Pyx_DECREF(__pyx_t_8); __pyx_t_8 = 0;
@@ -5967,7 +5965,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_6get_state(struct __pyx_obj_6mtr
   __pyx_t_1 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __Pyx_DECREF_SET(arrayObject_state, ((PyArrayObject *)__pyx_t_1));
+  __Pyx_DECREF(((PyObject *)arrayObject_state));
+  arrayObject_state = ((PyArrayObject *)__pyx_t_1);
   __pyx_t_1 = 0;
 
   /* "mtrand.pyx":647
@@ -6351,7 +6350,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_8set_state(struct __pyx_obj_6mtr
       __pyx_t_12 = __pyx_t_5;
       __Pyx_INCREF(__pyx_t_12);
       __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-      __Pyx_XDECREF_SET(arrayObject_obj, ((PyArrayObject *)__pyx_t_12));
+      __Pyx_XDECREF(((PyObject *)arrayObject_obj));
+      arrayObject_obj = ((PyArrayObject *)__pyx_t_12);
       __pyx_t_12 = 0;
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
       __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
@@ -7445,7 +7445,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-  __Pyx_DECREF_SET(__pyx_v_a, __pyx_t_4);
+  __Pyx_DECREF(__pyx_v_a);
+  __pyx_v_a = __pyx_t_4;
   __pyx_t_4 = 0;
 
   /* "mtrand.pyx":1008
@@ -7704,7 +7705,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
     __pyx_t_3 = __pyx_t_4;
     __Pyx_INCREF(__pyx_t_3);
     __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-    __Pyx_DECREF_SET(__pyx_v_p, __pyx_t_3);
+    __Pyx_DECREF(__pyx_v_p);
+    __pyx_v_p = __pyx_t_3;
     __pyx_t_3 = 0;
 
     /* "mtrand.pyx":1026
@@ -7906,7 +7908,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
     __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
     __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-    __Pyx_DECREF_SET(__pyx_v_size, __pyx_t_13);
+    __Pyx_DECREF(__pyx_v_size);
+    __pyx_v_size = __pyx_t_13;
     __pyx_t_13 = 0;
     goto __pyx_L21;
   }
@@ -7920,7 +7923,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
  *         # Actual sampling
  */
     __Pyx_INCREF(__pyx_int_1);
-    __Pyx_DECREF_SET(__pyx_v_size, __pyx_int_1);
+    __Pyx_DECREF(__pyx_v_size);
+    __pyx_v_size = __pyx_int_1;
   }
   __pyx_L21:;
 
@@ -7973,7 +7977,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
       __pyx_t_13 = __Pyx_PyNumber_InPlaceDivide(__pyx_v_cdf, __pyx_t_3); if (unlikely(!__pyx_t_13)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1047; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_13);
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-      __Pyx_DECREF_SET(__pyx_v_cdf, __pyx_t_13);
+      __Pyx_DECREF(__pyx_v_cdf);
+      __pyx_v_cdf = __pyx_t_13;
       __pyx_t_13 = 0;
 
       /* "mtrand.pyx":1048
@@ -8050,7 +8055,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
       __Pyx_DECREF(__pyx_t_13); __pyx_t_13 = 0;
       __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
       __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-      __Pyx_DECREF_SET(__pyx_v_idx, __pyx_t_4);
+      __Pyx_DECREF(__pyx_v_idx);
+      __pyx_v_idx = __pyx_t_4;
       __pyx_t_4 = 0;
       goto __pyx_L23;
     }
@@ -8195,7 +8201,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
       __pyx_t_13 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_empty_tuple), NULL); if (unlikely(!__pyx_t_13)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1062; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_13);
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-      __Pyx_DECREF_SET(__pyx_v_p, __pyx_t_13);
+      __Pyx_DECREF(__pyx_v_p);
+      __pyx_v_p = __pyx_t_13;
       __pyx_t_13 = 0;
 
       /* "mtrand.pyx":1063
@@ -8280,7 +8287,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
         __Pyx_GOTREF(__pyx_t_2);
         __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
         __Pyx_DECREF(((PyObject *)__pyx_t_13)); __pyx_t_13 = 0;
-        __Pyx_XDECREF_SET(__pyx_v_x, __pyx_t_2);
+        __Pyx_XDECREF(__pyx_v_x);
+        __pyx_v_x = __pyx_t_2;
         __pyx_t_2 = 0;
 
         /* "mtrand.pyx":1067
@@ -8331,7 +8339,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
         __Pyx_GOTREF(__pyx_t_1);
         __Pyx_DECREF(__pyx_t_13); __pyx_t_13 = 0;
         __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-        __Pyx_XDECREF_SET(__pyx_v_cdf, __pyx_t_1);
+        __Pyx_XDECREF(__pyx_v_cdf);
+        __pyx_v_cdf = __pyx_t_1;
         __pyx_t_1 = 0;
 
         /* "mtrand.pyx":1070
@@ -8346,7 +8355,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
         __pyx_t_2 = __Pyx_PyNumber_InPlaceDivide(__pyx_v_cdf, __pyx_t_1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1070; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
         __Pyx_GOTREF(__pyx_t_2);
         __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-        __Pyx_DECREF_SET(__pyx_v_cdf, __pyx_t_2);
+        __Pyx_DECREF(__pyx_v_cdf);
+        __pyx_v_cdf = __pyx_t_2;
         __pyx_t_2 = 0;
 
         /* "mtrand.pyx":1071
@@ -8371,7 +8381,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
         __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
         __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
         __Pyx_DECREF(((PyObject *)__pyx_t_13)); __pyx_t_13 = 0;
-        __Pyx_XDECREF_SET(__pyx_v_new, __pyx_t_3);
+        __Pyx_XDECREF(__pyx_v_new);
+        __pyx_v_new = __pyx_t_3;
         __pyx_t_3 = 0;
 
         /* "mtrand.pyx":1072
@@ -8453,9 +8464,11 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
           {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1072; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
           __pyx_L31_unpacking_done:;
         }
-        __Pyx_XDECREF_SET(__pyx_v__, __pyx_t_1);
+        __Pyx_XDECREF(__pyx_v__);
+        __pyx_v__ = __pyx_t_1;
         __pyx_t_1 = 0;
-        __Pyx_XDECREF_SET(__pyx_v_unique_indices, __pyx_t_3);
+        __Pyx_XDECREF(__pyx_v_unique_indices);
+        __pyx_v_unique_indices = __pyx_t_3;
         __pyx_t_3 = 0;
 
         /* "mtrand.pyx":1073
@@ -8490,7 +8503,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
         __Pyx_GOTREF(__pyx_t_1);
         __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
         __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-        __Pyx_DECREF_SET(__pyx_v_new, __pyx_t_1);
+        __Pyx_DECREF(__pyx_v_new);
+        __pyx_v_new = __pyx_t_1;
         __pyx_t_1 = 0;
 
         /* "mtrand.pyx":1075
@@ -8520,7 +8534,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
         __pyx_t_1 = PyNumber_InPlaceAdd(__pyx_v_n_uniq, __pyx_t_2); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1076; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
         __Pyx_GOTREF(__pyx_t_1);
         __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-        __Pyx_DECREF_SET(__pyx_v_n_uniq, __pyx_t_1);
+        __Pyx_DECREF(__pyx_v_n_uniq);
+        __pyx_v_n_uniq = __pyx_t_1;
         __pyx_t_1 = 0;
       }
 
@@ -8622,7 +8637,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_24choice(struct __pyx_obj_6mtran
     __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_k_tuple_38), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1085; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    __Pyx_DECREF_SET(__pyx_v_idx, __pyx_t_2);
+    __Pyx_DECREF(__pyx_v_idx);
+    __pyx_v_idx = __pyx_t_2;
     __pyx_t_2 = 0;
     goto __pyx_L33;
   }
@@ -9376,7 +9392,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_32random_integers(struct __pyx_o
  *         return self.randint(low, high+1, size)
  */
     __Pyx_INCREF(__pyx_v_low);
-    __Pyx_DECREF_SET(__pyx_v_high, __pyx_v_low);
+    __Pyx_DECREF(__pyx_v_high);
+    __pyx_v_high = __pyx_v_low;
 
     /* "mtrand.pyx":1366
  *         if high is None:
@@ -9386,7 +9403,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_32random_integers(struct __pyx_o
  * 
  */
     __Pyx_INCREF(__pyx_int_1);
-    __Pyx_DECREF_SET(__pyx_v_low, __pyx_int_1);
+    __Pyx_DECREF(__pyx_v_low);
+    __pyx_v_low = __pyx_int_1;
     goto __pyx_L3;
   }
   __pyx_L3:;
@@ -16774,7 +16792,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
  *                 raise ValueError("p < 0")
  *             elif fp > 1:             # <<<<<<<<<<<<<<
  *                 raise ValueError("p > 1")
- *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
+ *             elif np.isnan(fp):
  */
     __pyx_t_1 = ((__pyx_v_fp > 1.0) != 0);
     if (__pyx_t_1) {
@@ -16783,8 +16801,8 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
  *                 raise ValueError("p < 0")
  *             elif fp > 1:
  *                 raise ValueError("p > 1")             # <<<<<<<<<<<<<<
- *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
- * 
+ *             elif np.isnan(fp):
+ *                 raise ValueError("p is nan")
  */
       __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_140), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
@@ -16793,17 +16811,59 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
       {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
-    __pyx_L5:;
 
     /* "mtrand.pyx":3514
  *             elif fp > 1:
  *                 raise ValueError("p > 1")
+ *             elif np.isnan(fp):             # <<<<<<<<<<<<<<
+ *                 raise ValueError("p is nan")
+ *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
+ */
+    __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_2);
+    __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__isnan); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_3);
+    __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
+    __pyx_t_2 = PyFloat_FromDouble(__pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_2);
+    __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_4);
+    PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_2);
+    __Pyx_GIVEREF(__pyx_t_2);
+    __pyx_t_2 = 0;
+    __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_2);
+    __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+    __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
+    __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
+    if (__pyx_t_1) {
+
+      /* "mtrand.pyx":3515
+ *                 raise ValueError("p > 1")
+ *             elif np.isnan(fp):
+ *                 raise ValueError("p is nan")             # <<<<<<<<<<<<<<
+ *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
+ * 
+ */
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_142), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __Pyx_GOTREF(__pyx_t_2);
+      __Pyx_Raise(__pyx_t_2, 0, 0, 0);
+      __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      goto __pyx_L5;
+    }
+    __pyx_L5:;
+
+    /* "mtrand.pyx":3516
+ *             elif np.isnan(fp):
+ *                 raise ValueError("p is nan")
  *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)             # <<<<<<<<<<<<<<
  * 
  *         PyErr_Clear()
  */
     __Pyx_XDECREF(__pyx_r);
-    __pyx_t_2 = __pyx_f_6mtrand_discnp_array_sc(__pyx_v_self->internal_state, rk_binomial, __pyx_v_size, __pyx_v_ln, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __pyx_f_6mtrand_discnp_array_sc(__pyx_v_self->internal_state, rk_binomial, __pyx_v_size, __pyx_v_ln, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __pyx_r = __pyx_t_2;
     __pyx_t_2 = 0;
@@ -16812,7 +16872,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":3516
+  /* "mtrand.pyx":3518
  *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -16821,54 +16881,54 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":3518
+  /* "mtrand.pyx":3520
  *         PyErr_Clear()
  * 
  *         on = <ndarray>PyArray_FROM_OTF(n, NPY_LONG, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(n, 0)):
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_n, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_n, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __pyx_t_2;
-  __Pyx_INCREF(__pyx_t_3);
+  __pyx_t_4 = __pyx_t_2;
+  __Pyx_INCREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_v_on = ((PyArrayObject *)__pyx_t_3);
-  __pyx_t_3 = 0;
+  __pyx_v_on = ((PyArrayObject *)__pyx_t_4);
+  __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":3519
+  /* "mtrand.pyx":3521
  * 
  *         on = <ndarray>PyArray_FROM_OTF(n, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less(n, 0)):
  *             raise ValueError("n < 0")
  */
-  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __pyx_t_3;
+  __pyx_t_4 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_4);
+  __pyx_t_2 = __pyx_t_4;
   __Pyx_INCREF(__pyx_t_2);
-  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __pyx_v_op = ((PyArrayObject *)__pyx_t_2);
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":3520
+  /* "mtrand.pyx":3522
  *         on = <ndarray>PyArray_FROM_OTF(n, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(n, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("n < 0")
  *         if np.any(np.less(p, 0)):
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_3);
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_4);
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_v_n);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_v_n);
@@ -16876,57 +16936,57 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_2, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
+  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_t_5);
   __Pyx_GIVEREF(__pyx_t_5);
   __pyx_t_5 = 0;
-  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3521
+    /* "mtrand.pyx":3523
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(n, 0)):
  *             raise ValueError("n < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")
  */
-    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_141), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_143), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_Raise(__pyx_t_5, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L6;
   }
   __pyx_L6:;
 
-  /* "mtrand.pyx":3522
+  /* "mtrand.pyx":3524
  *         if np.any(np.less(n, 0)):
  *             raise ValueError("n < 0")
  *         if np.any(np.less(p, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):
  */
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__less); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_3);
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(__pyx_v_p);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_v_p);
@@ -16934,98 +16994,98 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_4 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_4);
-  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_3);
+  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_4);
-  __Pyx_GIVEREF(__pyx_t_4);
-  __pyx_t_4 = 0;
-  __pyx_t_4 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_4);
+  PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_3);
+  __Pyx_GIVEREF(__pyx_t_3);
+  __pyx_t_3 = 0;
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_4); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3523
+    /* "mtrand.pyx":3525
  *             raise ValueError("n < 0")
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")
  */
-    __pyx_t_4 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_142), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-    __Pyx_GOTREF(__pyx_t_4);
-    __Pyx_Raise(__pyx_t_4, 0, 0, 0);
-    __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_144), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_3);
+    __Pyx_Raise(__pyx_t_3, 0, 0, 0);
+    __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L7;
   }
   __pyx_L7:;
 
-  /* "mtrand.pyx":3524
+  /* "mtrand.pyx":3526
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p > 1")
  *         return discnp_array(self.internal_state, rk_binomial, size, on, op)
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_3);
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__greater); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_3);
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__greater); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_4);
+  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+  __pyx_t_3 = PyTuple_New(2); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_3);
   __Pyx_INCREF(__pyx_v_p);
-  PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_v_p);
+  PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_v_p);
   __Pyx_GIVEREF(__pyx_v_p);
   __Pyx_INCREF(__pyx_int_1);
-  PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_int_1);
+  PyTuple_SET_ITEM(__pyx_t_3, 1, __pyx_int_1);
   __Pyx_GIVEREF(__pyx_int_1);
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_3);
-  __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_3);
-  __Pyx_GIVEREF(__pyx_t_3);
-  __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
+  __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
+  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
+  PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_t_4);
+  __Pyx_GIVEREF(__pyx_t_4);
+  __pyx_t_4 = 0;
+  __pyx_t_4 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
+  __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_4); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3525
+    /* "mtrand.pyx":3527
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")             # <<<<<<<<<<<<<<
  *         return discnp_array(self.internal_state, rk_binomial, size, on, op)
  * 
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_143), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-    __Pyx_GOTREF(__pyx_t_3);
-    __Pyx_Raise(__pyx_t_3, 0, 0, 0);
-    __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_4 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_145), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __Pyx_GOTREF(__pyx_t_4);
+    __Pyx_Raise(__pyx_t_4, 0, 0, 0);
+    __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L8;
   }
   __pyx_L8:;
 
-  /* "mtrand.pyx":3526
+  /* "mtrand.pyx":3528
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")
  *         return discnp_array(self.internal_state, rk_binomial, size, on, op)             # <<<<<<<<<<<<<<
@@ -17033,10 +17093,10 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_82binomial(struct __pyx_obj_6mtr
  *     def negative_binomial(self, n, p, size=None):
  */
   __Pyx_XDECREF(__pyx_r);
-  __pyx_t_3 = __pyx_f_6mtrand_discnp_array(__pyx_v_self->internal_state, rk_binomial, __pyx_v_size, __pyx_v_on, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_t_3);
-  __pyx_r = __pyx_t_3;
-  __pyx_t_3 = 0;
+  __pyx_t_4 = __pyx_f_6mtrand_discnp_array(__pyx_v_self->internal_state, rk_binomial, __pyx_v_size, __pyx_v_on, __pyx_v_op); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_t_4);
+  __pyx_r = __pyx_t_4;
+  __pyx_t_4 = 0;
   goto __pyx_L0;
 
   __pyx_r = Py_None; __Pyx_INCREF(Py_None);
@@ -17073,7 +17133,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_85negative_binomial(PyObject *__
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__n,&__pyx_n_s__p,&__pyx_n_s__size,0};
     PyObject* values[3] = {0,0,0};
 
-    /* "mtrand.pyx":3528
+    /* "mtrand.pyx":3530
  *         return discnp_array(self.internal_state, rk_binomial, size, on, op)
  * 
  *     def negative_binomial(self, n, p, size=None):             # <<<<<<<<<<<<<<
@@ -17099,7 +17159,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_85negative_binomial(PyObject *__
         case  1:
         if (likely((values[1] = PyDict_GetItem(__pyx_kwds, __pyx_n_s__p)) != 0)) kw_args--;
         else {
-          __Pyx_RaiseArgtupleInvalid("negative_binomial", 0, 2, 3, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3528; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+          __Pyx_RaiseArgtupleInvalid("negative_binomial", 0, 2, 3, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3530; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
         }
         case  2:
         if (kw_args > 0) {
@@ -17108,7 +17168,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_85negative_binomial(PyObject *__
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "negative_binomial") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3528; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "negative_binomial") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3530; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -17125,7 +17185,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_85negative_binomial(PyObject *__
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("negative_binomial", 0, 2, 3, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3528; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("negative_binomial", 0, 2, 3, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3530; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.negative_binomial", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -17153,7 +17213,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("negative_binomial", 0);
 
-  /* "mtrand.pyx":3598
+  /* "mtrand.pyx":3600
  *         cdef double fp
  * 
  *         fp = PyFloat_AsDouble(p)             # <<<<<<<<<<<<<<
@@ -17162,7 +17222,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
  */
   __pyx_v_fp = PyFloat_AsDouble(__pyx_v_p);
 
-  /* "mtrand.pyx":3599
+  /* "mtrand.pyx":3601
  * 
  *         fp = PyFloat_AsDouble(p)
  *         fn = PyFloat_AsDouble(n)             # <<<<<<<<<<<<<<
@@ -17171,7 +17231,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
  */
   __pyx_v_fn = PyFloat_AsDouble(__pyx_v_n);
 
-  /* "mtrand.pyx":3600
+  /* "mtrand.pyx":3602
  *         fp = PyFloat_AsDouble(p)
  *         fn = PyFloat_AsDouble(n)
  *         if not PyErr_Occurred():             # <<<<<<<<<<<<<<
@@ -17181,7 +17241,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   __pyx_t_1 = ((!(PyErr_Occurred() != 0)) != 0);
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3601
+    /* "mtrand.pyx":3603
  *         fn = PyFloat_AsDouble(n)
  *         if not PyErr_Occurred():
  *             if fn <= 0:             # <<<<<<<<<<<<<<
@@ -17191,23 +17251,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
     __pyx_t_1 = ((__pyx_v_fn <= 0.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3602
+      /* "mtrand.pyx":3604
  *         if not PyErr_Occurred():
  *             if fn <= 0:
  *                 raise ValueError("n <= 0")             # <<<<<<<<<<<<<<
  *             if fp < 0:
  *                 raise ValueError("p < 0")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_145), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3602; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_147), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3604; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3602; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3604; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L4;
     }
     __pyx_L4:;
 
-    /* "mtrand.pyx":3603
+    /* "mtrand.pyx":3605
  *             if fn <= 0:
  *                 raise ValueError("n <= 0")
  *             if fp < 0:             # <<<<<<<<<<<<<<
@@ -17217,22 +17277,22 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
     __pyx_t_1 = ((__pyx_v_fp < 0.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3604
+      /* "mtrand.pyx":3606
  *                 raise ValueError("n <= 0")
  *             if fp < 0:
  *                 raise ValueError("p < 0")             # <<<<<<<<<<<<<<
  *             elif fp > 1:
  *                 raise ValueError("p > 1")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_146), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3604; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_148), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3606; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3604; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3606; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
 
-    /* "mtrand.pyx":3605
+    /* "mtrand.pyx":3607
  *             if fp < 0:
  *                 raise ValueError("p < 0")
  *             elif fp > 1:             # <<<<<<<<<<<<<<
@@ -17242,23 +17302,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
     __pyx_t_1 = ((__pyx_v_fp > 1.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3606
+      /* "mtrand.pyx":3608
  *                 raise ValueError("p < 0")
  *             elif fp > 1:
  *                 raise ValueError("p > 1")             # <<<<<<<<<<<<<<
  *             return discdd_array_sc(self.internal_state, rk_negative_binomial,
  *                                    size, fn, fp)
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_147), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3606; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_149), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3608; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3606; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3608; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
     __pyx_L5:;
 
-    /* "mtrand.pyx":3607
+    /* "mtrand.pyx":3609
  *             elif fp > 1:
  *                 raise ValueError("p > 1")
  *             return discdd_array_sc(self.internal_state, rk_negative_binomial,             # <<<<<<<<<<<<<<
@@ -17267,14 +17327,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
  */
     __Pyx_XDECREF(__pyx_r);
 
-    /* "mtrand.pyx":3608
+    /* "mtrand.pyx":3610
  *                 raise ValueError("p > 1")
  *             return discdd_array_sc(self.internal_state, rk_negative_binomial,
  *                                    size, fn, fp)             # <<<<<<<<<<<<<<
  * 
  *         PyErr_Clear()
  */
-    __pyx_t_2 = __pyx_f_6mtrand_discdd_array_sc(__pyx_v_self->internal_state, rk_negative_binomial, __pyx_v_size, __pyx_v_fn, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3607; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __pyx_f_6mtrand_discdd_array_sc(__pyx_v_self->internal_state, rk_negative_binomial, __pyx_v_size, __pyx_v_fn, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3609; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __pyx_r = __pyx_t_2;
     __pyx_t_2 = 0;
@@ -17283,7 +17343,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":3610
+  /* "mtrand.pyx":3612
  *                                    size, fn, fp)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -17292,14 +17352,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":3612
+  /* "mtrand.pyx":3614
  *         PyErr_Clear()
  * 
  *         on = <ndarray>PyArray_FROM_OTF(n, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(n, 0)):
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_n, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3612; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_n, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -17307,14 +17367,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   __pyx_v_on = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":3613
+  /* "mtrand.pyx":3615
  * 
  *         on = <ndarray>PyArray_FROM_OTF(n, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less_equal(n, 0)):
  *             raise ValueError("n <= 0")
  */
-  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3613; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3615; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_t_2 = __pyx_t_3;
   __Pyx_INCREF(__pyx_t_2);
@@ -17322,24 +17382,24 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   __pyx_v_op = ((PyArrayObject *)__pyx_t_2);
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":3614
+  /* "mtrand.pyx":3616
  *         on = <ndarray>PyArray_FROM_OTF(n, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(n, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("n <= 0")
  *         if np.any(np.less(p, 0)):
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less_equal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less_equal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_v_n);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_v_n);
@@ -17347,57 +17407,57 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_2, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_t_5);
   __Pyx_GIVEREF(__pyx_t_5);
   __pyx_t_5 = 0;
-  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3614; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3615
+    /* "mtrand.pyx":3617
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(n, 0)):
  *             raise ValueError("n <= 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")
  */
-    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_148), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3615; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_150), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3617; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_Raise(__pyx_t_5, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3615; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3617; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L6;
   }
   __pyx_L6:;
 
-  /* "mtrand.pyx":3616
+  /* "mtrand.pyx":3618
  *         if np.any(np.less_equal(n, 0)):
  *             raise ValueError("n <= 0")
  *         if np.any(np.less(p, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):
  */
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__less); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__less); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(__pyx_v_p);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_v_p);
@@ -17405,57 +17465,57 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_4 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_4);
   __Pyx_GIVEREF(__pyx_t_4);
   __pyx_t_4 = 0;
-  __pyx_t_4 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_4); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3616; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_4); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3617
+    /* "mtrand.pyx":3619
  *             raise ValueError("n <= 0")
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")
  */
-    __pyx_t_4 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_149), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3617; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_4 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_151), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3619; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_4);
     __Pyx_Raise(__pyx_t_4, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3617; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3619; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L7;
   }
   __pyx_L7:;
 
-  /* "mtrand.pyx":3618
+  /* "mtrand.pyx":3620
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p > 1")
  *         return discdd_array(self.internal_state, rk_negative_binomial, size,
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__greater); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__greater); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_INCREF(__pyx_v_p);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_v_p);
@@ -17463,40 +17523,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
   __Pyx_INCREF(__pyx_int_1);
   PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_int_1);
   __Pyx_GIVEREF(__pyx_int_1);
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3618; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3619
+    /* "mtrand.pyx":3621
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")             # <<<<<<<<<<<<<<
  *         return discdd_array(self.internal_state, rk_negative_binomial, size,
  *                             on, op)
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_150), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3619; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_152), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3621; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3619; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3621; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L8;
   }
   __pyx_L8:;
 
-  /* "mtrand.pyx":3620
+  /* "mtrand.pyx":3622
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")
  *         return discdd_array(self.internal_state, rk_negative_binomial, size,             # <<<<<<<<<<<<<<
@@ -17505,14 +17565,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_84negative_binomial(struct __pyx
  */
   __Pyx_XDECREF(__pyx_r);
 
-  /* "mtrand.pyx":3621
+  /* "mtrand.pyx":3623
  *             raise ValueError("p > 1")
  *         return discdd_array(self.internal_state, rk_negative_binomial, size,
  *                             on, op)             # <<<<<<<<<<<<<<
  * 
  *     def poisson(self, lam=1.0, size=None):
  */
-  __pyx_t_3 = __pyx_f_6mtrand_discdd_array(__pyx_v_self->internal_state, rk_negative_binomial, __pyx_v_size, __pyx_v_on, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3620; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __pyx_f_6mtrand_discdd_array(__pyx_v_self->internal_state, rk_negative_binomial, __pyx_v_size, __pyx_v_on, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3622; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_r = __pyx_t_3;
   __pyx_t_3 = 0;
@@ -17550,9 +17610,9 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_87poisson(PyObject *__pyx_v_self
   {
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__lam,&__pyx_n_s__size,0};
     PyObject* values[2] = {0,0};
-    values[0] = __pyx_k_151;
+    values[0] = __pyx_k_153;
 
-    /* "mtrand.pyx":3623
+    /* "mtrand.pyx":3625
  *                             on, op)
  * 
  *     def poisson(self, lam=1.0, size=None):             # <<<<<<<<<<<<<<
@@ -17583,7 +17643,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_87poisson(PyObject *__pyx_v_self
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "poisson") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3623; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "poisson") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3625; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -17598,7 +17658,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_87poisson(PyObject *__pyx_v_self
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("poisson", 0, 0, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3623; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("poisson", 0, 0, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3625; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.poisson", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -17624,7 +17684,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("poisson", 0);
 
-  /* "mtrand.pyx":3677
+  /* "mtrand.pyx":3679
  *         cdef ndarray olam
  *         cdef double flam
  *         flam = PyFloat_AsDouble(lam)             # <<<<<<<<<<<<<<
@@ -17633,7 +17693,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
  */
   __pyx_v_flam = PyFloat_AsDouble(__pyx_v_lam);
 
-  /* "mtrand.pyx":3678
+  /* "mtrand.pyx":3680
  *         cdef double flam
  *         flam = PyFloat_AsDouble(lam)
  *         if not PyErr_Occurred():             # <<<<<<<<<<<<<<
@@ -17643,66 +17703,66 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
   __pyx_t_1 = ((!(PyErr_Occurred() != 0)) != 0);
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3679
+    /* "mtrand.pyx":3681
  *         flam = PyFloat_AsDouble(lam)
  *         if not PyErr_Occurred():
  *             if lam < 0:             # <<<<<<<<<<<<<<
  *                 raise ValueError("lam < 0")
  *             if lam > self.poisson_lam_max:
  */
-    __pyx_t_2 = PyObject_RichCompare(__pyx_v_lam, __pyx_int_0, Py_LT); __Pyx_XGOTREF(__pyx_t_2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3679; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-    __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3679; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = PyObject_RichCompare(__pyx_v_lam, __pyx_int_0, Py_LT); __Pyx_XGOTREF(__pyx_t_2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3681; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3681; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3680
+      /* "mtrand.pyx":3682
  *         if not PyErr_Occurred():
  *             if lam < 0:
  *                 raise ValueError("lam < 0")             # <<<<<<<<<<<<<<
  *             if lam > self.poisson_lam_max:
  *                 raise ValueError("lam value too large")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_153), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3680; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_155), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3682; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3680; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3682; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L4;
     }
     __pyx_L4:;
 
-    /* "mtrand.pyx":3681
+    /* "mtrand.pyx":3683
  *             if lam < 0:
  *                 raise ValueError("lam < 0")
  *             if lam > self.poisson_lam_max:             # <<<<<<<<<<<<<<
  *                 raise ValueError("lam value too large")
  *             return discd_array_sc(self.internal_state, rk_poisson, size, flam)
  */
-    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__poisson_lam_max); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3681; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__poisson_lam_max); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3683; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
-    __pyx_t_3 = PyObject_RichCompare(__pyx_v_lam, __pyx_t_2, Py_GT); __Pyx_XGOTREF(__pyx_t_3); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3681; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_RichCompare(__pyx_v_lam, __pyx_t_2, Py_GT); __Pyx_XGOTREF(__pyx_t_3); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3683; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-    __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3681; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3683; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3682
+      /* "mtrand.pyx":3684
  *                 raise ValueError("lam < 0")
  *             if lam > self.poisson_lam_max:
  *                 raise ValueError("lam value too large")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_poisson, size, flam)
  * 
  */
-      __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_155), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3682; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_157), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3684; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_3);
       __Pyx_Raise(__pyx_t_3, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3682; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3684; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
     __pyx_L5:;
 
-    /* "mtrand.pyx":3683
+    /* "mtrand.pyx":3685
  *             if lam > self.poisson_lam_max:
  *                 raise ValueError("lam value too large")
  *             return discd_array_sc(self.internal_state, rk_poisson, size, flam)             # <<<<<<<<<<<<<<
@@ -17710,7 +17770,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
  *         PyErr_Clear()
  */
     __Pyx_XDECREF(__pyx_r);
-    __pyx_t_3 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_poisson, __pyx_v_size, __pyx_v_flam); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3683; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_poisson, __pyx_v_size, __pyx_v_flam); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3685; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __pyx_r = __pyx_t_3;
     __pyx_t_3 = 0;
@@ -17719,7 +17779,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":3685
+  /* "mtrand.pyx":3687
  *             return discd_array_sc(self.internal_state, rk_poisson, size, flam)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -17728,14 +17788,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":3687
+  /* "mtrand.pyx":3689
  *         PyErr_Clear()
  * 
  *         olam = <ndarray>PyArray_FROM_OTF(lam, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less(olam, 0)):
  *             raise ValueError("lam < 0")
  */
-  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_lam, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3687; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_lam, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3689; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_t_2 = __pyx_t_3;
   __Pyx_INCREF(__pyx_t_2);
@@ -17743,24 +17803,24 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
   __pyx_v_olam = ((PyArrayObject *)__pyx_t_2);
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":3688
+  /* "mtrand.pyx":3690
  * 
  *         olam = <ndarray>PyArray_FROM_OTF(lam, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(olam, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("lam < 0")
  *         if np.any(np.greater(olam, self.poisson_lam_max)):
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(((PyObject *)__pyx_v_olam));
   PyTuple_SET_ITEM(__pyx_t_2, 0, ((PyObject *)__pyx_v_olam));
@@ -17768,59 +17828,59 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_2, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_t_5);
   __Pyx_GIVEREF(__pyx_t_5);
   __pyx_t_5 = 0;
-  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3688; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3689
+    /* "mtrand.pyx":3691
  *         olam = <ndarray>PyArray_FROM_OTF(lam, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(olam, 0)):
  *             raise ValueError("lam < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(olam, self.poisson_lam_max)):
  *             raise ValueError("lam value too large.")
  */
-    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_156), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3689; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_158), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3691; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_Raise(__pyx_t_5, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3689; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3691; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L6;
   }
   __pyx_L6:;
 
-  /* "mtrand.pyx":3690
+  /* "mtrand.pyx":3692
  *         if np.any(np.less(olam, 0)):
  *             raise ValueError("lam < 0")
  *         if np.any(np.greater(olam, self.poisson_lam_max)):             # <<<<<<<<<<<<<<
  *             raise ValueError("lam value too large.")
  *         return discd_array(self.internal_state, rk_poisson, size, olam)
  */
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__greater); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__greater); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__poisson_lam_max); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__poisson_lam_max); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_INCREF(((PyObject *)__pyx_v_olam));
   PyTuple_SET_ITEM(__pyx_t_4, 0, ((PyObject *)__pyx_v_olam));
@@ -17828,40 +17888,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
   PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_t_5);
   __Pyx_GIVEREF(__pyx_t_5);
   __pyx_t_5 = 0;
-  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_5);
   __Pyx_GIVEREF(__pyx_t_5);
   __pyx_t_5 = 0;
-  __pyx_t_5 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3690; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3691
+    /* "mtrand.pyx":3693
  *             raise ValueError("lam < 0")
  *         if np.any(np.greater(olam, self.poisson_lam_max)):
  *             raise ValueError("lam value too large.")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_poisson, size, olam)
  * 
  */
-    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_158), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3691; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_160), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3693; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_Raise(__pyx_t_5, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3691; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3693; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L7;
   }
   __pyx_L7:;
 
-  /* "mtrand.pyx":3692
+  /* "mtrand.pyx":3694
  *         if np.any(np.greater(olam, self.poisson_lam_max)):
  *             raise ValueError("lam value too large.")
  *         return discd_array(self.internal_state, rk_poisson, size, olam)             # <<<<<<<<<<<<<<
@@ -17869,7 +17929,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_86poisson(struct __pyx_obj_6mtra
  *     def zipf(self, a, size=None):
  */
   __Pyx_XDECREF(__pyx_r);
-  __pyx_t_5 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_poisson, __pyx_v_size, __pyx_v_olam); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3692; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_poisson, __pyx_v_size, __pyx_v_olam); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3694; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __pyx_r = __pyx_t_5;
   __pyx_t_5 = 0;
@@ -17907,7 +17967,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_89zipf(PyObject *__pyx_v_self, P
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__a,&__pyx_n_s__size,0};
     PyObject* values[2] = {0,0};
 
-    /* "mtrand.pyx":3694
+    /* "mtrand.pyx":3696
  *         return discd_array(self.internal_state, rk_poisson, size, olam)
  * 
  *     def zipf(self, a, size=None):             # <<<<<<<<<<<<<<
@@ -17936,7 +17996,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_89zipf(PyObject *__pyx_v_self, P
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "zipf") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3694; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "zipf") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3696; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -17951,7 +18011,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_89zipf(PyObject *__pyx_v_self, P
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("zipf", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3694; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("zipf", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3696; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.zipf", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -17977,7 +18037,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("zipf", 0);
 
-  /* "mtrand.pyx":3769
+  /* "mtrand.pyx":3771
  *         cdef double fa
  * 
  *         fa = PyFloat_AsDouble(a)             # <<<<<<<<<<<<<<
@@ -17986,7 +18046,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
  */
   __pyx_v_fa = PyFloat_AsDouble(__pyx_v_a);
 
-  /* "mtrand.pyx":3770
+  /* "mtrand.pyx":3772
  * 
  *         fa = PyFloat_AsDouble(a)
  *         if not PyErr_Occurred():             # <<<<<<<<<<<<<<
@@ -17996,7 +18056,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
   __pyx_t_1 = ((!(PyErr_Occurred() != 0)) != 0);
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3771
+    /* "mtrand.pyx":3773
  *         fa = PyFloat_AsDouble(a)
  *         if not PyErr_Occurred():
  *             if fa <= 1.0:             # <<<<<<<<<<<<<<
@@ -18006,23 +18066,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
     __pyx_t_1 = ((__pyx_v_fa <= 1.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3772
+      /* "mtrand.pyx":3774
  *         if not PyErr_Occurred():
  *             if fa <= 1.0:
  *                 raise ValueError("a <= 1.0")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_zipf, size, fa)
  * 
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_160), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3772; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_162), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3774; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3772; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3774; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L4;
     }
     __pyx_L4:;
 
-    /* "mtrand.pyx":3773
+    /* "mtrand.pyx":3775
  *             if fa <= 1.0:
  *                 raise ValueError("a <= 1.0")
  *             return discd_array_sc(self.internal_state, rk_zipf, size, fa)             # <<<<<<<<<<<<<<
@@ -18030,7 +18090,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
  *         PyErr_Clear()
  */
     __Pyx_XDECREF(__pyx_r);
-    __pyx_t_2 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_zipf, __pyx_v_size, __pyx_v_fa); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3773; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_zipf, __pyx_v_size, __pyx_v_fa); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3775; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __pyx_r = __pyx_t_2;
     __pyx_t_2 = 0;
@@ -18039,7 +18099,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":3775
+  /* "mtrand.pyx":3777
  *             return discd_array_sc(self.internal_state, rk_zipf, size, fa)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -18048,14 +18108,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":3777
+  /* "mtrand.pyx":3779
  *         PyErr_Clear()
  * 
  *         oa = <ndarray>PyArray_FROM_OTF(a, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less_equal(oa, 1.0)):
  *             raise ValueError("a <= 1.0")
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_a, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3777; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_a, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3779; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -18063,26 +18123,26 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
   __pyx_v_oa = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":3778
+  /* "mtrand.pyx":3780
  * 
  *         oa = <ndarray>PyArray_FROM_OTF(a, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(oa, 1.0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("a <= 1.0")
  *         return discd_array(self.internal_state, rk_zipf, size, oa)
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less_equal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less_equal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(((PyObject *)__pyx_v_oa));
   PyTuple_SET_ITEM(__pyx_t_5, 0, ((PyObject *)__pyx_v_oa));
@@ -18090,40 +18150,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3778; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3779
+    /* "mtrand.pyx":3781
  *         oa = <ndarray>PyArray_FROM_OTF(a, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(oa, 1.0)):
  *             raise ValueError("a <= 1.0")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_zipf, size, oa)
  * 
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_161), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3779; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_163), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3781; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3779; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3781; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L5;
   }
   __pyx_L5:;
 
-  /* "mtrand.pyx":3780
+  /* "mtrand.pyx":3782
  *         if np.any(np.less_equal(oa, 1.0)):
  *             raise ValueError("a <= 1.0")
  *         return discd_array(self.internal_state, rk_zipf, size, oa)             # <<<<<<<<<<<<<<
@@ -18131,7 +18191,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_88zipf(struct __pyx_obj_6mtrand_
  *     def geometric(self, p, size=None):
  */
   __Pyx_XDECREF(__pyx_r);
-  __pyx_t_3 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_zipf, __pyx_v_size, __pyx_v_oa); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3780; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_zipf, __pyx_v_size, __pyx_v_oa); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3782; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_r = __pyx_t_3;
   __pyx_t_3 = 0;
@@ -18169,7 +18229,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_91geometric(PyObject *__pyx_v_se
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__p,&__pyx_n_s__size,0};
     PyObject* values[2] = {0,0};
 
-    /* "mtrand.pyx":3782
+    /* "mtrand.pyx":3784
  *         return discd_array(self.internal_state, rk_zipf, size, oa)
  * 
  *     def geometric(self, p, size=None):             # <<<<<<<<<<<<<<
@@ -18198,7 +18258,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_91geometric(PyObject *__pyx_v_se
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "geometric") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3782; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "geometric") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3784; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -18213,7 +18273,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_91geometric(PyObject *__pyx_v_se
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("geometric", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3782; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("geometric", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3784; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.geometric", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -18239,7 +18299,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("geometric", 0);
 
-  /* "mtrand.pyx":3830
+  /* "mtrand.pyx":3832
  *         cdef double fp
  * 
  *         fp = PyFloat_AsDouble(p)             # <<<<<<<<<<<<<<
@@ -18248,7 +18308,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
  */
   __pyx_v_fp = PyFloat_AsDouble(__pyx_v_p);
 
-  /* "mtrand.pyx":3831
+  /* "mtrand.pyx":3833
  * 
  *         fp = PyFloat_AsDouble(p)
  *         if not PyErr_Occurred():             # <<<<<<<<<<<<<<
@@ -18258,7 +18318,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
   __pyx_t_1 = ((!(PyErr_Occurred() != 0)) != 0);
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3832
+    /* "mtrand.pyx":3834
  *         fp = PyFloat_AsDouble(p)
  *         if not PyErr_Occurred():
  *             if fp < 0.0:             # <<<<<<<<<<<<<<
@@ -18268,23 +18328,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
     __pyx_t_1 = ((__pyx_v_fp < 0.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3833
+      /* "mtrand.pyx":3835
  *         if not PyErr_Occurred():
  *             if fp < 0.0:
  *                 raise ValueError("p < 0.0")             # <<<<<<<<<<<<<<
  *             if fp > 1.0:
  *                 raise ValueError("p > 1.0")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_163), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3833; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_165), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3835; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3833; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3835; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L4;
     }
     __pyx_L4:;
 
-    /* "mtrand.pyx":3834
+    /* "mtrand.pyx":3836
  *             if fp < 0.0:
  *                 raise ValueError("p < 0.0")
  *             if fp > 1.0:             # <<<<<<<<<<<<<<
@@ -18294,23 +18354,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
     __pyx_t_1 = ((__pyx_v_fp > 1.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3835
+      /* "mtrand.pyx":3837
  *                 raise ValueError("p < 0.0")
  *             if fp > 1.0:
  *                 raise ValueError("p > 1.0")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_geometric, size, fp)
  * 
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_165), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3835; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_167), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3837; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3835; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3837; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
     __pyx_L5:;
 
-    /* "mtrand.pyx":3836
+    /* "mtrand.pyx":3838
  *             if fp > 1.0:
  *                 raise ValueError("p > 1.0")
  *             return discd_array_sc(self.internal_state, rk_geometric, size, fp)             # <<<<<<<<<<<<<<
@@ -18318,7 +18378,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
  *         PyErr_Clear()
  */
     __Pyx_XDECREF(__pyx_r);
-    __pyx_t_2 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_geometric, __pyx_v_size, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3836; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_geometric, __pyx_v_size, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3838; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __pyx_r = __pyx_t_2;
     __pyx_t_2 = 0;
@@ -18327,7 +18387,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":3838
+  /* "mtrand.pyx":3840
  *             return discd_array_sc(self.internal_state, rk_geometric, size, fp)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -18336,14 +18396,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":3841
+  /* "mtrand.pyx":3843
  * 
  * 
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less(op, 0.0)):
  *             raise ValueError("p < 0.0")
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3841; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3843; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -18351,26 +18411,26 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
   __pyx_v_op = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":3842
+  /* "mtrand.pyx":3844
  * 
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(op, 0.0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p < 0.0")
  *         if np.any(np.greater(op, 1.0)):
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyFloat_FromDouble(0.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyFloat_FromDouble(0.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(((PyObject *)__pyx_v_op));
   PyTuple_SET_ITEM(__pyx_t_5, 0, ((PyObject *)__pyx_v_op));
@@ -18378,59 +18438,59 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3842; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3843
+    /* "mtrand.pyx":3845
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(op, 0.0)):
  *             raise ValueError("p < 0.0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(op, 1.0)):
  *             raise ValueError("p > 1.0")
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_166), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3843; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_168), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3845; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3843; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3845; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L6;
   }
   __pyx_L6:;
 
-  /* "mtrand.pyx":3844
+  /* "mtrand.pyx":3846
  *         if np.any(np.less(op, 0.0)):
  *             raise ValueError("p < 0.0")
  *         if np.any(np.greater(op, 1.0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p > 1.0")
  *         return discd_array(self.internal_state, rk_geometric, size, op)
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__greater); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__greater); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_INCREF(((PyObject *)__pyx_v_op));
   PyTuple_SET_ITEM(__pyx_t_4, 0, ((PyObject *)__pyx_v_op));
@@ -18438,40 +18498,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
   PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3844; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3845
+    /* "mtrand.pyx":3847
  *             raise ValueError("p < 0.0")
  *         if np.any(np.greater(op, 1.0)):
  *             raise ValueError("p > 1.0")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_geometric, size, op)
  * 
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_167), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3845; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_169), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3847; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3845; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3847; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L7;
   }
   __pyx_L7:;
 
-  /* "mtrand.pyx":3846
+  /* "mtrand.pyx":3848
  *         if np.any(np.greater(op, 1.0)):
  *             raise ValueError("p > 1.0")
  *         return discd_array(self.internal_state, rk_geometric, size, op)             # <<<<<<<<<<<<<<
@@ -18479,7 +18539,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_90geometric(struct __pyx_obj_6mt
  *     def hypergeometric(self, ngood, nbad, nsample, size=None):
  */
   __Pyx_XDECREF(__pyx_r);
-  __pyx_t_3 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_geometric, __pyx_v_size, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3846; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_geometric, __pyx_v_size, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3848; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_r = __pyx_t_3;
   __pyx_t_3 = 0;
@@ -18519,7 +18579,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_93hypergeometric(PyObject *__pyx
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__ngood,&__pyx_n_s__nbad,&__pyx_n_s__nsample,&__pyx_n_s__size,0};
     PyObject* values[4] = {0,0,0,0};
 
-    /* "mtrand.pyx":3848
+    /* "mtrand.pyx":3850
  *         return discd_array(self.internal_state, rk_geometric, size, op)
  * 
  *     def hypergeometric(self, ngood, nbad, nsample, size=None):             # <<<<<<<<<<<<<<
@@ -18546,12 +18606,12 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_93hypergeometric(PyObject *__pyx
         case  1:
         if (likely((values[1] = PyDict_GetItem(__pyx_kwds, __pyx_n_s__nbad)) != 0)) kw_args--;
         else {
-          __Pyx_RaiseArgtupleInvalid("hypergeometric", 0, 3, 4, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3848; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+          __Pyx_RaiseArgtupleInvalid("hypergeometric", 0, 3, 4, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3850; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
         }
         case  2:
         if (likely((values[2] = PyDict_GetItem(__pyx_kwds, __pyx_n_s__nsample)) != 0)) kw_args--;
         else {
-          __Pyx_RaiseArgtupleInvalid("hypergeometric", 0, 3, 4, 2); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3848; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+          __Pyx_RaiseArgtupleInvalid("hypergeometric", 0, 3, 4, 2); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3850; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
         }
         case  3:
         if (kw_args > 0) {
@@ -18560,7 +18620,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_93hypergeometric(PyObject *__pyx
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "hypergeometric") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3848; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "hypergeometric") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3850; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -18579,7 +18639,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_93hypergeometric(PyObject *__pyx
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("hypergeometric", 0, 3, 4, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3848; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("hypergeometric", 0, 3, 4, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3850; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.hypergeometric", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -18610,7 +18670,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("hypergeometric", 0);
 
-  /* "mtrand.pyx":3936
+  /* "mtrand.pyx":3938
  *         cdef long lngood, lnbad, lnsample
  * 
  *         lngood = PyInt_AsLong(ngood)             # <<<<<<<<<<<<<<
@@ -18619,7 +18679,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
  */
   __pyx_v_lngood = PyInt_AsLong(__pyx_v_ngood);
 
-  /* "mtrand.pyx":3937
+  /* "mtrand.pyx":3939
  * 
  *         lngood = PyInt_AsLong(ngood)
  *         lnbad = PyInt_AsLong(nbad)             # <<<<<<<<<<<<<<
@@ -18628,7 +18688,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
  */
   __pyx_v_lnbad = PyInt_AsLong(__pyx_v_nbad);
 
-  /* "mtrand.pyx":3938
+  /* "mtrand.pyx":3940
  *         lngood = PyInt_AsLong(ngood)
  *         lnbad = PyInt_AsLong(nbad)
  *         lnsample = PyInt_AsLong(nsample)             # <<<<<<<<<<<<<<
@@ -18637,7 +18697,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
  */
   __pyx_v_lnsample = PyInt_AsLong(__pyx_v_nsample);
 
-  /* "mtrand.pyx":3939
+  /* "mtrand.pyx":3941
  *         lnbad = PyInt_AsLong(nbad)
  *         lnsample = PyInt_AsLong(nsample)
  *         if not PyErr_Occurred():             # <<<<<<<<<<<<<<
@@ -18647,7 +18707,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __pyx_t_1 = ((!(PyErr_Occurred() != 0)) != 0);
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3940
+    /* "mtrand.pyx":3942
  *         lnsample = PyInt_AsLong(nsample)
  *         if not PyErr_Occurred():
  *             if lngood < 0:             # <<<<<<<<<<<<<<
@@ -18657,23 +18717,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
     __pyx_t_1 = ((__pyx_v_lngood < 0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3941
+      /* "mtrand.pyx":3943
  *         if not PyErr_Occurred():
  *             if lngood < 0:
  *                 raise ValueError("ngood < 0")             # <<<<<<<<<<<<<<
  *             if lnbad < 0:
  *                 raise ValueError("nbad < 0")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_169), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3941; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_171), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3943; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3941; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3943; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L4;
     }
     __pyx_L4:;
 
-    /* "mtrand.pyx":3942
+    /* "mtrand.pyx":3944
  *             if lngood < 0:
  *                 raise ValueError("ngood < 0")
  *             if lnbad < 0:             # <<<<<<<<<<<<<<
@@ -18683,23 +18743,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
     __pyx_t_1 = ((__pyx_v_lnbad < 0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3943
+      /* "mtrand.pyx":3945
  *                 raise ValueError("ngood < 0")
  *             if lnbad < 0:
  *                 raise ValueError("nbad < 0")             # <<<<<<<<<<<<<<
  *             if lnsample < 1:
  *                 raise ValueError("nsample < 1")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_171), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3943; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_173), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3945; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3943; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3945; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
     __pyx_L5:;
 
-    /* "mtrand.pyx":3944
+    /* "mtrand.pyx":3946
  *             if lnbad < 0:
  *                 raise ValueError("nbad < 0")
  *             if lnsample < 1:             # <<<<<<<<<<<<<<
@@ -18709,23 +18769,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
     __pyx_t_1 = ((__pyx_v_lnsample < 1) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3945
+      /* "mtrand.pyx":3947
  *                 raise ValueError("nbad < 0")
  *             if lnsample < 1:
  *                 raise ValueError("nsample < 1")             # <<<<<<<<<<<<<<
  *             if lngood + lnbad < lnsample:
  *                 raise ValueError("ngood + nbad < nsample")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_173), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3945; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_175), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3947; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3945; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3947; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L6;
     }
     __pyx_L6:;
 
-    /* "mtrand.pyx":3946
+    /* "mtrand.pyx":3948
  *             if lnsample < 1:
  *                 raise ValueError("nsample < 1")
  *             if lngood + lnbad < lnsample:             # <<<<<<<<<<<<<<
@@ -18735,23 +18795,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
     __pyx_t_1 = (((__pyx_v_lngood + __pyx_v_lnbad) < __pyx_v_lnsample) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":3947
+      /* "mtrand.pyx":3949
  *                 raise ValueError("nsample < 1")
  *             if lngood + lnbad < lnsample:
  *                 raise ValueError("ngood + nbad < nsample")             # <<<<<<<<<<<<<<
  *             return discnmN_array_sc(self.internal_state, rk_hypergeometric, size,
  *                                     lngood, lnbad, lnsample)
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_175), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3947; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_177), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3949; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3947; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3949; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L7;
     }
     __pyx_L7:;
 
-    /* "mtrand.pyx":3948
+    /* "mtrand.pyx":3950
  *             if lngood + lnbad < lnsample:
  *                 raise ValueError("ngood + nbad < nsample")
  *             return discnmN_array_sc(self.internal_state, rk_hypergeometric, size,             # <<<<<<<<<<<<<<
@@ -18760,14 +18820,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
  */
     __Pyx_XDECREF(__pyx_r);
 
-    /* "mtrand.pyx":3949
+    /* "mtrand.pyx":3951
  *                 raise ValueError("ngood + nbad < nsample")
  *             return discnmN_array_sc(self.internal_state, rk_hypergeometric, size,
  *                                     lngood, lnbad, lnsample)             # <<<<<<<<<<<<<<
  * 
  *         PyErr_Clear()
  */
-    __pyx_t_2 = __pyx_f_6mtrand_discnmN_array_sc(__pyx_v_self->internal_state, rk_hypergeometric, __pyx_v_size, __pyx_v_lngood, __pyx_v_lnbad, __pyx_v_lnsample); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3948; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __pyx_f_6mtrand_discnmN_array_sc(__pyx_v_self->internal_state, rk_hypergeometric, __pyx_v_size, __pyx_v_lngood, __pyx_v_lnbad, __pyx_v_lnsample); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3950; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __pyx_r = __pyx_t_2;
     __pyx_t_2 = 0;
@@ -18776,7 +18836,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":3951
+  /* "mtrand.pyx":3953
  *                                     lngood, lnbad, lnsample)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -18785,14 +18845,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":3953
+  /* "mtrand.pyx":3955
  *         PyErr_Clear()
  * 
  *         ongood = <ndarray>PyArray_FROM_OTF(ngood, NPY_LONG, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         onbad = <ndarray>PyArray_FROM_OTF(nbad, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         onsample = <ndarray>PyArray_FROM_OTF(nsample, NPY_LONG, NPY_ARRAY_ALIGNED)
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_ngood, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3953; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_ngood, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3955; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -18800,14 +18860,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __pyx_v_ongood = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":3954
+  /* "mtrand.pyx":3956
  * 
  *         ongood = <ndarray>PyArray_FROM_OTF(ngood, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         onbad = <ndarray>PyArray_FROM_OTF(nbad, NPY_LONG, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         onsample = <ndarray>PyArray_FROM_OTF(nsample, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(ongood, 0)):
  */
-  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_nbad, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3954; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyArray_FROM_OTF(__pyx_v_nbad, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_t_2 = __pyx_t_3;
   __Pyx_INCREF(__pyx_t_2);
@@ -18815,14 +18875,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __pyx_v_onbad = ((PyArrayObject *)__pyx_t_2);
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":3955
+  /* "mtrand.pyx":3957
  *         ongood = <ndarray>PyArray_FROM_OTF(ngood, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         onbad = <ndarray>PyArray_FROM_OTF(nbad, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         onsample = <ndarray>PyArray_FROM_OTF(nsample, NPY_LONG, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less(ongood, 0)):
  *             raise ValueError("ngood < 0")
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_nsample, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3955; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_nsample, NPY_LONG, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3957; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -18830,24 +18890,24 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __pyx_v_onsample = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":3956
+  /* "mtrand.pyx":3958
  *         onbad = <ndarray>PyArray_FROM_OTF(nbad, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         onsample = <ndarray>PyArray_FROM_OTF(nsample, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(ongood, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("ngood < 0")
  *         if np.any(np.less(onbad, 0)):
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyTuple_New(2); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(2); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_INCREF(((PyObject *)__pyx_v_ongood));
   PyTuple_SET_ITEM(__pyx_t_3, 0, ((PyObject *)__pyx_v_ongood));
@@ -18855,57 +18915,57 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_3, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_t_5);
   __Pyx_GIVEREF(__pyx_t_5);
   __pyx_t_5 = 0;
-  __pyx_t_5 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3956; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_5); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3957
+    /* "mtrand.pyx":3959
  *         onsample = <ndarray>PyArray_FROM_OTF(nsample, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(ongood, 0)):
  *             raise ValueError("ngood < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(onbad, 0)):
  *             raise ValueError("nbad < 0")
  */
-    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_176), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3957; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_178), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3959; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_Raise(__pyx_t_5, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3957; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3959; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L8;
   }
   __pyx_L8:;
 
-  /* "mtrand.pyx":3958
+  /* "mtrand.pyx":3960
  *         if np.any(np.less(ongood, 0)):
  *             raise ValueError("ngood < 0")
  *         if np.any(np.less(onbad, 0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("nbad < 0")
  *         if np.any(np.less(onsample, 1)):
  */
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__less); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__less); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(((PyObject *)__pyx_v_onbad));
   PyTuple_SET_ITEM(__pyx_t_5, 0, ((PyObject *)__pyx_v_onbad));
@@ -18913,57 +18973,57 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __Pyx_INCREF(__pyx_int_0);
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_int_0);
   __Pyx_GIVEREF(__pyx_int_0);
-  __pyx_t_4 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_4);
   __Pyx_GIVEREF(__pyx_t_4);
   __pyx_t_4 = 0;
-  __pyx_t_4 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_4); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3958; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_4); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3959
+    /* "mtrand.pyx":3961
  *             raise ValueError("ngood < 0")
  *         if np.any(np.less(onbad, 0)):
  *             raise ValueError("nbad < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(onsample, 1)):
  *             raise ValueError("nsample < 1")
  */
-    __pyx_t_4 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_177), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3959; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_4 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_179), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3961; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_4);
     __Pyx_Raise(__pyx_t_4, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3959; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3961; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L9;
   }
   __pyx_L9:;
 
-  /* "mtrand.pyx":3960
+  /* "mtrand.pyx":3962
  *         if np.any(np.less(onbad, 0)):
  *             raise ValueError("nbad < 0")
  *         if np.any(np.less(onsample, 1)):             # <<<<<<<<<<<<<<
  *             raise ValueError("nsample < 1")
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__less); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__less); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_INCREF(((PyObject *)__pyx_v_onsample));
   PyTuple_SET_ITEM(__pyx_t_4, 0, ((PyObject *)__pyx_v_onsample));
@@ -18971,62 +19031,62 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __Pyx_INCREF(__pyx_int_1);
   PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_int_1);
   __Pyx_GIVEREF(__pyx_int_1);
-  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_2);
   __Pyx_GIVEREF(__pyx_t_2);
   __pyx_t_2 = 0;
-  __pyx_t_2 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3960; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3961
+    /* "mtrand.pyx":3963
  *             raise ValueError("nbad < 0")
  *         if np.any(np.less(onsample, 1)):
  *             raise ValueError("nsample < 1")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):
  *             raise ValueError("ngood + nbad < nsample")
  */
-    __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_178), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3961; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_180), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3963; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_Raise(__pyx_t_2, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3961; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3963; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L10;
   }
   __pyx_L10:;
 
-  /* "mtrand.pyx":3962
+  /* "mtrand.pyx":3964
  *         if np.any(np.less(onsample, 1)):
  *             raise ValueError("nsample < 1")
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):             # <<<<<<<<<<<<<<
  *             raise ValueError("ngood + nbad < nsample")
  *         return discnmN_array(self.internal_state, rk_hypergeometric, size,
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__less); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__add); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__add); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(((PyObject *)__pyx_v_ongood));
   PyTuple_SET_ITEM(__pyx_t_2, 0, ((PyObject *)__pyx_v_ongood));
@@ -19034,11 +19094,11 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   __Pyx_INCREF(((PyObject *)__pyx_v_onbad));
   PyTuple_SET_ITEM(__pyx_t_2, 1, ((PyObject *)__pyx_v_onbad));
   __Pyx_GIVEREF(((PyObject *)__pyx_v_onbad));
-  __pyx_t_6 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_6 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_6);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_t_6);
   __Pyx_GIVEREF(__pyx_t_6);
@@ -19046,40 +19106,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
   PyTuple_SET_ITEM(__pyx_t_2, 1, ((PyObject *)__pyx_v_onsample));
   __Pyx_GIVEREF(((PyObject *)__pyx_v_onsample));
   __pyx_t_6 = 0;
-  __pyx_t_6 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_6 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_6);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_t_6);
   __Pyx_GIVEREF(__pyx_t_6);
   __pyx_t_6 = 0;
-  __pyx_t_6 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_6 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_6);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_6); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3962; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_6); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_6); __pyx_t_6 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":3963
+    /* "mtrand.pyx":3965
  *             raise ValueError("nsample < 1")
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):
  *             raise ValueError("ngood + nbad < nsample")             # <<<<<<<<<<<<<<
  *         return discnmN_array(self.internal_state, rk_hypergeometric, size,
  *             ongood, onbad, onsample)
  */
-    __pyx_t_6 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_179), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3963; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_6 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_181), NULL); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3965; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_6);
     __Pyx_Raise(__pyx_t_6, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_6); __pyx_t_6 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3963; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3965; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L11;
   }
   __pyx_L11:;
 
-  /* "mtrand.pyx":3964
+  /* "mtrand.pyx":3966
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):
  *             raise ValueError("ngood + nbad < nsample")
  *         return discnmN_array(self.internal_state, rk_hypergeometric, size,             # <<<<<<<<<<<<<<
@@ -19088,14 +19148,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_92hypergeometric(struct __pyx_ob
  */
   __Pyx_XDECREF(__pyx_r);
 
-  /* "mtrand.pyx":3965
+  /* "mtrand.pyx":3967
  *             raise ValueError("ngood + nbad < nsample")
  *         return discnmN_array(self.internal_state, rk_hypergeometric, size,
  *             ongood, onbad, onsample)             # <<<<<<<<<<<<<<
  * 
  *     def logseries(self, p, size=None):
  */
-  __pyx_t_6 = __pyx_f_6mtrand_discnmN_array(__pyx_v_self->internal_state, rk_hypergeometric, __pyx_v_size, __pyx_v_ongood, __pyx_v_onbad, __pyx_v_onsample); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3964; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_6 = __pyx_f_6mtrand_discnmN_array(__pyx_v_self->internal_state, rk_hypergeometric, __pyx_v_size, __pyx_v_ongood, __pyx_v_onbad, __pyx_v_onsample); if (unlikely(!__pyx_t_6)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3966; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_6);
   __pyx_r = __pyx_t_6;
   __pyx_t_6 = 0;
@@ -19136,7 +19196,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_95logseries(PyObject *__pyx_v_se
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__p,&__pyx_n_s__size,0};
     PyObject* values[2] = {0,0};
 
-    /* "mtrand.pyx":3967
+    /* "mtrand.pyx":3969
  *             ongood, onbad, onsample)
  * 
  *     def logseries(self, p, size=None):             # <<<<<<<<<<<<<<
@@ -19165,7 +19225,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_95logseries(PyObject *__pyx_v_se
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "logseries") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3967; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "logseries") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3969; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -19180,7 +19240,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_95logseries(PyObject *__pyx_v_se
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("logseries", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3967; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("logseries", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3969; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.logseries", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -19206,7 +19266,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("logseries", 0);
 
-  /* "mtrand.pyx":4044
+  /* "mtrand.pyx":4046
  *         cdef double fp
  * 
  *         fp = PyFloat_AsDouble(p)             # <<<<<<<<<<<<<<
@@ -19215,7 +19275,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
  */
   __pyx_v_fp = PyFloat_AsDouble(__pyx_v_p);
 
-  /* "mtrand.pyx":4045
+  /* "mtrand.pyx":4047
  * 
  *         fp = PyFloat_AsDouble(p)
  *         if not PyErr_Occurred():             # <<<<<<<<<<<<<<
@@ -19225,7 +19285,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
   __pyx_t_1 = ((!(PyErr_Occurred() != 0)) != 0);
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":4046
+    /* "mtrand.pyx":4048
  *         fp = PyFloat_AsDouble(p)
  *         if not PyErr_Occurred():
  *             if fp <= 0.0:             # <<<<<<<<<<<<<<
@@ -19235,23 +19295,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
     __pyx_t_1 = ((__pyx_v_fp <= 0.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":4047
+      /* "mtrand.pyx":4049
  *         if not PyErr_Occurred():
  *             if fp <= 0.0:
  *                 raise ValueError("p <= 0.0")             # <<<<<<<<<<<<<<
  *             if fp >= 1.0:
  *                 raise ValueError("p >= 1.0")
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_181), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4047; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_183), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4049; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4047; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4049; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L4;
     }
     __pyx_L4:;
 
-    /* "mtrand.pyx":4048
+    /* "mtrand.pyx":4050
  *             if fp <= 0.0:
  *                 raise ValueError("p <= 0.0")
  *             if fp >= 1.0:             # <<<<<<<<<<<<<<
@@ -19261,23 +19321,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
     __pyx_t_1 = ((__pyx_v_fp >= 1.0) != 0);
     if (__pyx_t_1) {
 
-      /* "mtrand.pyx":4049
+      /* "mtrand.pyx":4051
  *                 raise ValueError("p <= 0.0")
  *             if fp >= 1.0:
  *                 raise ValueError("p >= 1.0")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_logseries, size, fp)
  * 
  */
-      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_183), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4049; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_185), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4051; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
       __Pyx_Raise(__pyx_t_2, 0, 0, 0);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4049; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4051; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       goto __pyx_L5;
     }
     __pyx_L5:;
 
-    /* "mtrand.pyx":4050
+    /* "mtrand.pyx":4052
  *             if fp >= 1.0:
  *                 raise ValueError("p >= 1.0")
  *             return discd_array_sc(self.internal_state, rk_logseries, size, fp)             # <<<<<<<<<<<<<<
@@ -19285,7 +19345,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
  *         PyErr_Clear()
  */
     __Pyx_XDECREF(__pyx_r);
-    __pyx_t_2 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_logseries, __pyx_v_size, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4050; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __pyx_f_6mtrand_discd_array_sc(__pyx_v_self->internal_state, rk_logseries, __pyx_v_size, __pyx_v_fp); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4052; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __pyx_r = __pyx_t_2;
     __pyx_t_2 = 0;
@@ -19294,7 +19354,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":4052
+  /* "mtrand.pyx":4054
  *             return discd_array_sc(self.internal_state, rk_logseries, size, fp)
  * 
  *         PyErr_Clear()             # <<<<<<<<<<<<<<
@@ -19303,14 +19363,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
  */
   PyErr_Clear();
 
-  /* "mtrand.pyx":4054
+  /* "mtrand.pyx":4056
  *         PyErr_Clear()
  * 
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)             # <<<<<<<<<<<<<<
  *         if np.any(np.less_equal(op, 0.0)):
  *             raise ValueError("p <= 0.0")
  */
-  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4054; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_FROM_OTF(__pyx_v_p, NPY_DOUBLE, NPY_ARRAY_ALIGNED); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4056; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -19318,26 +19378,26 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
   __pyx_v_op = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4055
+  /* "mtrand.pyx":4057
  * 
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(op, 0.0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p <= 0.0")
  *         if np.any(np.greater_equal(op, 1.0)):
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less_equal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__less_equal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyFloat_FromDouble(0.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyFloat_FromDouble(0.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(((PyObject *)__pyx_v_op));
   PyTuple_SET_ITEM(__pyx_t_5, 0, ((PyObject *)__pyx_v_op));
@@ -19345,59 +19405,59 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4055; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":4056
+    /* "mtrand.pyx":4058
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(op, 0.0)):
  *             raise ValueError("p <= 0.0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater_equal(op, 1.0)):
  *             raise ValueError("p >= 1.0")
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_184), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4056; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_186), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4058; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4056; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4058; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L6;
   }
   __pyx_L6:;
 
-  /* "mtrand.pyx":4057
+  /* "mtrand.pyx":4059
  *         if np.any(np.less_equal(op, 0.0)):
  *             raise ValueError("p <= 0.0")
  *         if np.any(np.greater_equal(op, 1.0)):             # <<<<<<<<<<<<<<
  *             raise ValueError("p >= 1.0")
  *         return discd_array(self.internal_state, rk_logseries, size, op)
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__any); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__greater_equal); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__greater_equal); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_INCREF(((PyObject *)__pyx_v_op));
   PyTuple_SET_ITEM(__pyx_t_4, 0, ((PyObject *)__pyx_v_op));
@@ -19405,40 +19465,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
   PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(1); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
-  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4057; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_IsTrue(__pyx_t_3); if (unlikely(__pyx_t_1 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_1) {
 
-    /* "mtrand.pyx":4058
+    /* "mtrand.pyx":4060
  *             raise ValueError("p <= 0.0")
  *         if np.any(np.greater_equal(op, 1.0)):
  *             raise ValueError("p >= 1.0")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_logseries, size, op)
  * 
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_185), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4058; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_187), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4060; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4058; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4060; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L7;
   }
   __pyx_L7:;
 
-  /* "mtrand.pyx":4059
+  /* "mtrand.pyx":4061
  *         if np.any(np.greater_equal(op, 1.0)):
  *             raise ValueError("p >= 1.0")
  *         return discd_array(self.internal_state, rk_logseries, size, op)             # <<<<<<<<<<<<<<
@@ -19446,7 +19506,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_94logseries(struct __pyx_obj_6mt
  *     # Multivariate distributions:
  */
   __Pyx_XDECREF(__pyx_r);
-  __pyx_t_3 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_logseries, __pyx_v_size, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4059; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __pyx_f_6mtrand_discd_array(__pyx_v_self->internal_state, rk_logseries, __pyx_v_size, __pyx_v_op); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4061; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __pyx_r = __pyx_t_3;
   __pyx_t_3 = 0;
@@ -19485,7 +19545,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_97multivariate_normal(PyObject *
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__mean,&__pyx_n_s__cov,&__pyx_n_s__size,0};
     PyObject* values[3] = {0,0,0};
 
-    /* "mtrand.pyx":4062
+    /* "mtrand.pyx":4064
  * 
  *     # Multivariate distributions:
  *     def multivariate_normal(self, mean, cov, size=None):             # <<<<<<<<<<<<<<
@@ -19511,7 +19571,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_97multivariate_normal(PyObject *
         case  1:
         if (likely((values[1] = PyDict_GetItem(__pyx_kwds, __pyx_n_s__cov)) != 0)) kw_args--;
         else {
-          __Pyx_RaiseArgtupleInvalid("multivariate_normal", 0, 2, 3, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4062; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+          __Pyx_RaiseArgtupleInvalid("multivariate_normal", 0, 2, 3, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4064; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
         }
         case  2:
         if (kw_args > 0) {
@@ -19520,7 +19580,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_97multivariate_normal(PyObject *
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "multivariate_normal") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4062; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "multivariate_normal") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4064; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -19537,7 +19597,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_97multivariate_normal(PyObject *
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("multivariate_normal", 0, 2, 3, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4062; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("multivariate_normal", 0, 2, 3, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4064; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.multivariate_normal", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -19577,77 +19637,79 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   __Pyx_INCREF(__pyx_v_mean);
   __Pyx_INCREF(__pyx_v_cov);
 
-  /* "mtrand.pyx":4153
+  /* "mtrand.pyx":4155
  * 
  *         """
  *         from numpy.dual import svd             # <<<<<<<<<<<<<<
  * 
  *         # Check preconditions on arguments
  */
-  __pyx_t_1 = PyList_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4153; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyList_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4155; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_INCREF(((PyObject *)__pyx_n_s__svd));
   PyList_SET_ITEM(__pyx_t_1, 0, ((PyObject *)__pyx_n_s__svd));
   __Pyx_GIVEREF(((PyObject *)__pyx_n_s__svd));
-  __pyx_t_2 = __Pyx_Import(((PyObject *)__pyx_n_s_186), ((PyObject *)__pyx_t_1), -1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4153; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_Import(((PyObject *)__pyx_n_s_188), ((PyObject *)__pyx_t_1), -1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4155; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
-  __pyx_t_1 = __Pyx_ImportFrom(__pyx_t_2, __pyx_n_s__svd); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4153; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_ImportFrom(__pyx_t_2, __pyx_n_s__svd); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4155; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_INCREF(__pyx_t_1);
   __pyx_v_svd = __pyx_t_1;
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":4156
+  /* "mtrand.pyx":4158
  * 
  *         # Check preconditions on arguments
  *         mean = np.array(mean)             # <<<<<<<<<<<<<<
  *         cov = np.array(cov)
  *         if size is None:
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4156; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4158; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__array); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4156; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__array); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4158; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4156; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4158; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_v_mean);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_v_mean);
   __Pyx_GIVEREF(__pyx_v_mean);
-  __pyx_t_3 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4156; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4158; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __Pyx_DECREF_SET(__pyx_v_mean, __pyx_t_3);
+  __Pyx_DECREF(__pyx_v_mean);
+  __pyx_v_mean = __pyx_t_3;
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4157
+  /* "mtrand.pyx":4159
  *         # Check preconditions on arguments
  *         mean = np.array(mean)
  *         cov = np.array(cov)             # <<<<<<<<<<<<<<
  *         if size is None:
  *             shape = []
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4157; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4159; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__array); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4157; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__array); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4159; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4157; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4159; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_INCREF(__pyx_v_cov);
   PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_v_cov);
   __Pyx_GIVEREF(__pyx_v_cov);
-  __pyx_t_1 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4157; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4159; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-  __Pyx_DECREF_SET(__pyx_v_cov, __pyx_t_1);
+  __Pyx_DECREF(__pyx_v_cov);
+  __pyx_v_cov = __pyx_t_1;
   __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4158
+  /* "mtrand.pyx":4160
  *         mean = np.array(mean)
  *         cov = np.array(cov)
  *         if size is None:             # <<<<<<<<<<<<<<
@@ -19658,33 +19720,33 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   __pyx_t_5 = (__pyx_t_4 != 0);
   if (__pyx_t_5) {
 
-    /* "mtrand.pyx":4159
+    /* "mtrand.pyx":4161
  *         cov = np.array(cov)
  *         if size is None:
  *             shape = []             # <<<<<<<<<<<<<<
  *         elif isinstance(size, (int, long, np.integer)):
  *             shape = [size]
  */
-    __pyx_t_1 = PyList_New(0); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4159; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyList_New(0); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4161; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __pyx_v_shape = ((PyObject *)__pyx_t_1);
     __pyx_t_1 = 0;
     goto __pyx_L3;
   }
 
-  /* "mtrand.pyx":4160
+  /* "mtrand.pyx":4162
  *         if size is None:
  *             shape = []
  *         elif isinstance(size, (int, long, np.integer)):             # <<<<<<<<<<<<<<
  *             shape = [size]
  *         else:
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4160; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4162; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__integer); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4160; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__integer); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4162; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_1 = PyTuple_New(3); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4160; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyTuple_New(3); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4162; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_INCREF(((PyObject *)((PyObject*)(&PyInt_Type))));
   PyTuple_SET_ITEM(__pyx_t_1, 0, ((PyObject *)((PyObject*)(&PyInt_Type))));
@@ -19695,19 +19757,19 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   PyTuple_SET_ITEM(__pyx_t_1, 2, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_5 = PyObject_IsInstance(__pyx_v_size, ((PyObject *)__pyx_t_1)); if (unlikely(__pyx_t_5 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4160; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyObject_IsInstance(__pyx_v_size, ((PyObject *)__pyx_t_1)); if (unlikely(__pyx_t_5 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4162; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
   __pyx_t_4 = (__pyx_t_5 != 0);
   if (__pyx_t_4) {
 
-    /* "mtrand.pyx":4161
+    /* "mtrand.pyx":4163
  *             shape = []
  *         elif isinstance(size, (int, long, np.integer)):
  *             shape = [size]             # <<<<<<<<<<<<<<
  *         else:
  *             shape = size
  */
-    __pyx_t_1 = PyList_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4161; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyList_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4163; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __Pyx_INCREF(__pyx_v_size);
     PyList_SET_ITEM(__pyx_t_1, 0, __pyx_v_size);
@@ -19718,7 +19780,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   }
   /*else*/ {
 
-    /* "mtrand.pyx":4163
+    /* "mtrand.pyx":4165
  *             shape = [size]
  *         else:
  *             shape = size             # <<<<<<<<<<<<<<
@@ -19730,63 +19792,63 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":4165
+  /* "mtrand.pyx":4167
  *             shape = size
  * 
  *         if len(mean.shape) != 1:             # <<<<<<<<<<<<<<
  *                raise ValueError("mean must be 1 dimensional")
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):
  */
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4165; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_6 = PyObject_Length(__pyx_t_1); if (unlikely(__pyx_t_6 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4165; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_6 = PyObject_Length(__pyx_t_1); if (unlikely(__pyx_t_6 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __pyx_t_4 = ((__pyx_t_6 != 1) != 0);
   if (__pyx_t_4) {
 
-    /* "mtrand.pyx":4166
+    /* "mtrand.pyx":4168
  * 
  *         if len(mean.shape) != 1:
  *                raise ValueError("mean must be 1 dimensional")             # <<<<<<<<<<<<<<
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):
  *                raise ValueError("cov must be 2 dimensional and square")
  */
-    __pyx_t_1 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_188), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4166; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_190), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4168; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __Pyx_Raise(__pyx_t_1, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4166; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4168; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L4;
   }
   __pyx_L4:;
 
-  /* "mtrand.pyx":4167
+  /* "mtrand.pyx":4169
  *         if len(mean.shape) != 1:
  *                raise ValueError("mean must be 1 dimensional")
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):             # <<<<<<<<<<<<<<
  *                raise ValueError("cov must be 2 dimensional and square")
  *         if mean.shape[0] != cov.shape[0]:
  */
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_6 = PyObject_Length(__pyx_t_1); if (unlikely(__pyx_t_6 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_6 = PyObject_Length(__pyx_t_1); if (unlikely(__pyx_t_6 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __pyx_t_4 = (__pyx_t_6 != 2);
   if (!__pyx_t_4) {
-    __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-    __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    __pyx_t_2 = __Pyx_GetItemInt(__pyx_t_1, 1, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_2) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __Pyx_GetItemInt(__pyx_t_1, 1, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_2) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-    __pyx_t_1 = PyObject_RichCompare(__pyx_t_3, __pyx_t_2, Py_NE); __Pyx_XGOTREF(__pyx_t_1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyObject_RichCompare(__pyx_t_3, __pyx_t_2, Py_NE); __Pyx_XGOTREF(__pyx_t_1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-    __pyx_t_5 = __Pyx_PyObject_IsTrue(__pyx_t_1); if (unlikely(__pyx_t_5 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4167; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = __Pyx_PyObject_IsTrue(__pyx_t_1); if (unlikely(__pyx_t_5 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
     __pyx_t_7 = __pyx_t_5;
   } else {
@@ -19794,124 +19856,124 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   }
   if (__pyx_t_7) {
 
-    /* "mtrand.pyx":4168
+    /* "mtrand.pyx":4170
  *                raise ValueError("mean must be 1 dimensional")
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):
  *                raise ValueError("cov must be 2 dimensional and square")             # <<<<<<<<<<<<<<
  *         if mean.shape[0] != cov.shape[0]:
  *                raise ValueError("mean and cov must have same length")
  */
-    __pyx_t_1 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_190), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4168; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_192), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4170; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __Pyx_Raise(__pyx_t_1, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4168; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4170; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L5;
   }
   __pyx_L5:;
 
-  /* "mtrand.pyx":4169
+  /* "mtrand.pyx":4171
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):
  *                raise ValueError("cov must be 2 dimensional and square")
  *         if mean.shape[0] != cov.shape[0]:             # <<<<<<<<<<<<<<
  *                raise ValueError("mean and cov must have same length")
  * 
  */
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4171; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_2 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_2) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_2) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4171; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_cov, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4171; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4171; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_1 = PyObject_RichCompare(__pyx_t_2, __pyx_t_3, Py_NE); __Pyx_XGOTREF(__pyx_t_1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_RichCompare(__pyx_t_2, __pyx_t_3, Py_NE); __Pyx_XGOTREF(__pyx_t_1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4171; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_7 = __Pyx_PyObject_IsTrue(__pyx_t_1); if (unlikely(__pyx_t_7 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4169; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_7 = __Pyx_PyObject_IsTrue(__pyx_t_1); if (unlikely(__pyx_t_7 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4171; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   if (__pyx_t_7) {
 
-    /* "mtrand.pyx":4170
+    /* "mtrand.pyx":4172
  *                raise ValueError("cov must be 2 dimensional and square")
  *         if mean.shape[0] != cov.shape[0]:
  *                raise ValueError("mean and cov must have same length")             # <<<<<<<<<<<<<<
  * 
  *         # Compute shape of output and create a matrix of independent
  */
-    __pyx_t_1 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_192), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4170; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_194), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4172; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __Pyx_Raise(__pyx_t_1, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4170; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4172; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L6;
   }
   __pyx_L6:;
 
-  /* "mtrand.pyx":4176
+  /* "mtrand.pyx":4178
  *         # with the same length as mean and as many rows are necessary to
  *         # form a matrix of shape final_shape.
  *         final_shape = list(shape[:])             # <<<<<<<<<<<<<<
  *         final_shape.append(mean.shape[0])
  *         x = self.standard_normal(final_shape).reshape(-1, mean.shape[0])
  */
-  __pyx_t_1 = __Pyx_PyObject_GetSlice(__pyx_v_shape, 0, 0, NULL, NULL, &__pyx_k_slice_193, 0, 0, 1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4176; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetSlice(__pyx_v_shape, 0, 0, NULL, NULL, &__pyx_k_slice_195, 0, 0, 1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4176; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_t_1);
   __Pyx_GIVEREF(__pyx_t_1);
   __pyx_t_1 = 0;
-  __pyx_t_1 = PyObject_Call(((PyObject *)((PyObject*)(&PyList_Type))), ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4176; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Call(((PyObject *)((PyObject*)(&PyList_Type))), ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
   __pyx_v_final_shape = ((PyObject*)__pyx_t_1);
   __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4177
+  /* "mtrand.pyx":4179
  *         # form a matrix of shape final_shape.
  *         final_shape = list(shape[:])
  *         final_shape.append(mean.shape[0])             # <<<<<<<<<<<<<<
  *         x = self.standard_normal(final_shape).reshape(-1, mean.shape[0])
  * 
  */
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4177; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4179; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4177; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_1, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4179; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_8 = __Pyx_PyList_Append(__pyx_v_final_shape, __pyx_t_3); if (unlikely(__pyx_t_8 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4177; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_8 = __Pyx_PyList_Append(__pyx_v_final_shape, __pyx_t_3); if (unlikely(__pyx_t_8 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4179; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4178
+  /* "mtrand.pyx":4180
  *         final_shape = list(shape[:])
  *         final_shape.append(mean.shape[0])
  *         x = self.standard_normal(final_shape).reshape(-1, mean.shape[0])             # <<<<<<<<<<<<<<
  * 
  *         # Transform matrix of standard normals into matrix where each row
  */
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__standard_normal); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__standard_normal); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_1 = PyTuple_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyTuple_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_INCREF(((PyObject *)__pyx_v_final_shape));
   PyTuple_SET_ITEM(__pyx_t_1, 0, ((PyObject *)__pyx_v_final_shape));
   __Pyx_GIVEREF(((PyObject *)__pyx_v_final_shape));
-  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_1), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_1), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__reshape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__reshape); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_v_mean, __pyx_n_s__shape); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_2, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetItemInt(__pyx_t_2, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_int_neg_1);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_int_neg_1);
@@ -19919,26 +19981,26 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   PyTuple_SET_ITEM(__pyx_t_2, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4180; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
   __pyx_v_x = __pyx_t_3;
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4193
+  /* "mtrand.pyx":4195
  *         # order to preserve current outputs. Note that symmetry has not
  *         # been checked.
  *         (u, s, v) = svd(cov)             # <<<<<<<<<<<<<<
  *         neg = (np.sum(u.T * v, axis=1) < 0) & (s > 0)
  *         if np.any(neg):
  */
-  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_INCREF(__pyx_v_cov);
   PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_v_cov);
   __Pyx_GIVEREF(__pyx_v_cov);
-  __pyx_t_2 = PyObject_Call(__pyx_v_svd, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyObject_Call(__pyx_v_svd, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
   if ((likely(PyTuple_CheckExact(__pyx_t_2))) || (PyList_CheckExact(__pyx_t_2))) {
@@ -19951,7 +20013,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
     if (unlikely(size != 3)) {
       if (size > 3) __Pyx_RaiseTooManyValuesError(3);
       else if (size >= 0) __Pyx_RaiseNeedMoreValuesError(size);
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     }
     #if CYTHON_COMPILING_IN_CPYTHON
     if (likely(PyTuple_CheckExact(sequence))) {
@@ -19967,18 +20029,18 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
     __Pyx_INCREF(__pyx_t_1);
     __Pyx_INCREF(__pyx_t_9);
     #else
-    __pyx_t_3 = PySequence_ITEM(sequence, 0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PySequence_ITEM(sequence, 0); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
-    __pyx_t_1 = PySequence_ITEM(sequence, 1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PySequence_ITEM(sequence, 1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    __pyx_t_9 = PySequence_ITEM(sequence, 2); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_9 = PySequence_ITEM(sequence, 2); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_9);
     #endif
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   } else
   {
     Py_ssize_t index = -1;
-    __pyx_t_10 = PyObject_GetIter(__pyx_t_2); if (unlikely(!__pyx_t_10)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_10 = PyObject_GetIter(__pyx_t_2); if (unlikely(!__pyx_t_10)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_10);
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
     __pyx_t_11 = Py_TYPE(__pyx_t_10)->tp_iternext;
@@ -19988,7 +20050,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
     __Pyx_GOTREF(__pyx_t_1);
     index = 2; __pyx_t_9 = __pyx_t_11(__pyx_t_10); if (unlikely(!__pyx_t_9)) goto __pyx_L7_unpacking_failed;
     __Pyx_GOTREF(__pyx_t_9);
-    if (__Pyx_IternextUnpackEndCheck(__pyx_t_11(__pyx_t_10), 3) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    if (__Pyx_IternextUnpackEndCheck(__pyx_t_11(__pyx_t_10), 3) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __pyx_t_11 = NULL;
     __Pyx_DECREF(__pyx_t_10); __pyx_t_10 = 0;
     goto __pyx_L8_unpacking_done;
@@ -19996,7 +20058,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
     __Pyx_DECREF(__pyx_t_10); __pyx_t_10 = 0;
     __pyx_t_11 = NULL;
     if (__Pyx_IterFinish() == 0) __Pyx_RaiseNeedMoreValuesError(index);
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4193; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __pyx_L8_unpacking_done:;
   }
   __pyx_v_u = __pyx_t_3;
@@ -20006,112 +20068,112 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   __pyx_v_v = __pyx_t_9;
   __pyx_t_9 = 0;
 
-  /* "mtrand.pyx":4194
+  /* "mtrand.pyx":4196
  *         # been checked.
  *         (u, s, v) = svd(cov)
  *         neg = (np.sum(u.T * v, axis=1) < 0) & (s > 0)             # <<<<<<<<<<<<<<
  *         if np.any(neg):
  *             s[neg] = 0.
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_9 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__sum); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_9 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__sum); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_9);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_v_u, __pyx_n_s__T); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_v_u, __pyx_n_s__T); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_1 = PyNumber_Multiply(__pyx_t_2, __pyx_v_v); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyNumber_Multiply(__pyx_t_2, __pyx_v_v); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_t_1);
   __Pyx_GIVEREF(__pyx_t_1);
   __pyx_t_1 = 0;
-  __pyx_t_1 = PyDict_New(); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyDict_New(); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(((PyObject *)__pyx_t_1));
-  if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_n_s__axis), __pyx_int_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __pyx_t_3 = PyObject_Call(__pyx_t_9, ((PyObject *)__pyx_t_2), ((PyObject *)__pyx_t_1)); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_n_s__axis), __pyx_int_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_9, ((PyObject *)__pyx_t_2), ((PyObject *)__pyx_t_1)); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_9); __pyx_t_9 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
-  __pyx_t_1 = PyObject_RichCompare(__pyx_t_3, __pyx_int_0, Py_LT); __Pyx_XGOTREF(__pyx_t_1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_RichCompare(__pyx_t_3, __pyx_int_0, Py_LT); __Pyx_XGOTREF(__pyx_t_1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_RichCompare(__pyx_v_s, __pyx_int_0, Py_GT); __Pyx_XGOTREF(__pyx_t_3); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __pyx_t_2 = PyNumber_And(__pyx_t_1, __pyx_t_3); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4194; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_RichCompare(__pyx_v_s, __pyx_int_0, Py_GT); __Pyx_XGOTREF(__pyx_t_3); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyNumber_And(__pyx_t_1, __pyx_t_3); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __pyx_v_neg = __pyx_t_2;
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":4195
+  /* "mtrand.pyx":4197
  *         (u, s, v) = svd(cov)
  *         neg = (np.sum(u.T * v, axis=1) < 0) & (s > 0)
  *         if np.any(neg):             # <<<<<<<<<<<<<<
  *             s[neg] = 0.
  *             warnings.warn("covariance is not positive-semidefinite.",
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__any); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_v_neg);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_v_neg);
   __Pyx_GIVEREF(__pyx_v_neg);
-  __pyx_t_1 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
-  __pyx_t_7 = __Pyx_PyObject_IsTrue(__pyx_t_1); if (unlikely(__pyx_t_7 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4195; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_7 = __Pyx_PyObject_IsTrue(__pyx_t_1); if (unlikely(__pyx_t_7 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   if (__pyx_t_7) {
 
-    /* "mtrand.pyx":4196
+    /* "mtrand.pyx":4198
  *         neg = (np.sum(u.T * v, axis=1) < 0) & (s > 0)
  *         if np.any(neg):
  *             s[neg] = 0.             # <<<<<<<<<<<<<<
  *             warnings.warn("covariance is not positive-semidefinite.",
  *                           RuntimeWarning)
  */
-    __pyx_t_1 = PyFloat_FromDouble(0.); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyFloat_FromDouble(0.); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4198; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    if (PyObject_SetItem(__pyx_v_s, __pyx_v_neg, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4196; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    if (PyObject_SetItem(__pyx_v_s, __pyx_v_neg, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4198; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-    /* "mtrand.pyx":4197
+    /* "mtrand.pyx":4199
  *         if np.any(neg):
  *             s[neg] = 0.
  *             warnings.warn("covariance is not positive-semidefinite.",             # <<<<<<<<<<<<<<
  *                           RuntimeWarning)
  * 
  */
-    __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__warnings); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__warnings); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4199; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__warn); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__warn); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4199; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-    /* "mtrand.pyx":4198
+    /* "mtrand.pyx":4200
  *             s[neg] = 0.
  *             warnings.warn("covariance is not positive-semidefinite.",
  *                           RuntimeWarning)             # <<<<<<<<<<<<<<
  * 
  *         x = np.dot(x, np.sqrt(s)[:, None] * v)
  */
-    __pyx_t_1 = PyTuple_New(2); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyTuple_New(2); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4199; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    __Pyx_INCREF(((PyObject *)__pyx_kp_s_194));
-    PyTuple_SET_ITEM(__pyx_t_1, 0, ((PyObject *)__pyx_kp_s_194));
-    __Pyx_GIVEREF(((PyObject *)__pyx_kp_s_194));
+    __Pyx_INCREF(((PyObject *)__pyx_kp_s_196));
+    PyTuple_SET_ITEM(__pyx_t_1, 0, ((PyObject *)__pyx_kp_s_196));
+    __Pyx_GIVEREF(((PyObject *)__pyx_kp_s_196));
     __Pyx_INCREF(__pyx_builtin_RuntimeWarning);
     PyTuple_SET_ITEM(__pyx_t_1, 1, __pyx_builtin_RuntimeWarning);
     __Pyx_GIVEREF(__pyx_builtin_RuntimeWarning);
-    __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_1), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4197; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_1), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4199; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
     __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
@@ -20120,39 +20182,39 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   }
   __pyx_L9:;
 
-  /* "mtrand.pyx":4200
+  /* "mtrand.pyx":4202
  *                           RuntimeWarning)
  * 
  *         x = np.dot(x, np.sqrt(s)[:, None] * v)             # <<<<<<<<<<<<<<
  *         x += mean
  *         x.shape = tuple(final_shape)
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__dot); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__dot); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__sqrt); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__sqrt); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(1); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_INCREF(__pyx_v_s);
   PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_v_s);
   __Pyx_GIVEREF(__pyx_v_s);
-  __pyx_t_9 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_9 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_9);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_GetItem(__pyx_t_9, ((PyObject *)__pyx_k_tuple_196)); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_GetItem(__pyx_t_9, ((PyObject *)__pyx_k_tuple_198)); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_9); __pyx_t_9 = 0;
-  __pyx_t_9 = PyNumber_Multiply(__pyx_t_3, __pyx_v_v); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_9 = PyNumber_Multiply(__pyx_t_3, __pyx_v_v); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_9);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-  __pyx_t_3 = PyTuple_New(2); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyTuple_New(2); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_INCREF(__pyx_v_x);
   PyTuple_SET_ITEM(__pyx_t_3, 0, __pyx_v_x);
@@ -20160,38 +20222,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_96multivariate_normal(struct __p
   PyTuple_SET_ITEM(__pyx_t_3, 1, __pyx_t_9);
   __Pyx_GIVEREF(__pyx_t_9);
   __pyx_t_9 = 0;
-  __pyx_t_9 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_9 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_3), NULL); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_9);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_3)); __pyx_t_3 = 0;
-  __Pyx_DECREF_SET(__pyx_v_x, __pyx_t_9);
+  __Pyx_DECREF(__pyx_v_x);
+  __pyx_v_x = __pyx_t_9;
   __pyx_t_9 = 0;
 
-  /* "mtrand.pyx":4201
+  /* "mtrand.pyx":4203
  * 
  *         x = np.dot(x, np.sqrt(s)[:, None] * v)
  *         x += mean             # <<<<<<<<<<<<<<
  *         x.shape = tuple(final_shape)
  *         return x
  */
-  __pyx_t_9 = PyNumber_InPlaceAdd(__pyx_v_x, __pyx_v_mean); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4201; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_9 = PyNumber_InPlaceAdd(__pyx_v_x, __pyx_v_mean); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4203; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_9);
-  __Pyx_DECREF_SET(__pyx_v_x, __pyx_t_9);
+  __Pyx_DECREF(__pyx_v_x);
+  __pyx_v_x = __pyx_t_9;
   __pyx_t_9 = 0;
 
-  /* "mtrand.pyx":4202
+  /* "mtrand.pyx":4204
  *         x = np.dot(x, np.sqrt(s)[:, None] * v)
  *         x += mean
  *         x.shape = tuple(final_shape)             # <<<<<<<<<<<<<<
  *         return x
  * 
  */
-  __pyx_t_9 = ((PyObject *)PyList_AsTuple(__pyx_v_final_shape)); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_9 = ((PyObject *)PyList_AsTuple(__pyx_v_final_shape)); if (unlikely(!__pyx_t_9)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4204; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(((PyObject *)__pyx_t_9));
-  if (__Pyx_PyObject_SetAttrStr(__pyx_v_x, __pyx_n_s__shape, ((PyObject *)__pyx_t_9)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (__Pyx_PyObject_SetAttrStr(__pyx_v_x, __pyx_n_s__shape, ((PyObject *)__pyx_t_9)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4204; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(((PyObject *)__pyx_t_9)); __pyx_t_9 = 0;
 
-  /* "mtrand.pyx":4203
+  /* "mtrand.pyx":4205
  *         x += mean
  *         x.shape = tuple(final_shape)
  *         return x             # <<<<<<<<<<<<<<
@@ -20246,7 +20310,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_99multinomial(PyObject *__pyx_v_
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__n,&__pyx_n_s__pvals,&__pyx_n_s__size,0};
     PyObject* values[3] = {0,0,0};
 
-    /* "mtrand.pyx":4205
+    /* "mtrand.pyx":4207
  *         return x
  * 
  *     def multinomial(self, npy_intp n, object pvals, size=None):             # <<<<<<<<<<<<<<
@@ -20272,7 +20336,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_99multinomial(PyObject *__pyx_v_
         case  1:
         if (likely((values[1] = PyDict_GetItem(__pyx_kwds, __pyx_n_s__pvals)) != 0)) kw_args--;
         else {
-          __Pyx_RaiseArgtupleInvalid("multinomial", 0, 2, 3, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4205; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+          __Pyx_RaiseArgtupleInvalid("multinomial", 0, 2, 3, 1); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4207; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
         }
         case  2:
         if (kw_args > 0) {
@@ -20281,7 +20345,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_99multinomial(PyObject *__pyx_v_
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "multinomial") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4205; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "multinomial") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4207; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -20292,13 +20356,13 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_99multinomial(PyObject *__pyx_v_
         default: goto __pyx_L5_argtuple_error;
       }
     }
-    __pyx_v_n = __Pyx_PyInt_from_py_npy_intp(values[0]); if (unlikely((__pyx_v_n == (npy_intp)-1) && PyErr_Occurred())) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4205; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+    __pyx_v_n = __Pyx_PyInt_from_py_npy_intp(values[0]); if (unlikely((__pyx_v_n == (npy_intp)-1) && PyErr_Occurred())) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4207; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
     __pyx_v_pvals = values[1];
     __pyx_v_size = values[2];
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("multinomial", 0, 2, 3, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4205; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("multinomial", 0, 2, 3, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4207; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.multinomial", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -20334,24 +20398,24 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("multinomial", 0);
 
-  /* "mtrand.pyx":4264
+  /* "mtrand.pyx":4266
  *         cdef double Sum
  * 
  *         d = len(pvals)             # <<<<<<<<<<<<<<
  *         parr = <ndarray>PyArray_ContiguousFromObject(pvals, NPY_DOUBLE, 1, 1)
  *         pix = <double*>PyArray_DATA(parr)
  */
-  __pyx_t_1 = PyObject_Length(__pyx_v_pvals); if (unlikely(__pyx_t_1 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4264; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Length(__pyx_v_pvals); if (unlikely(__pyx_t_1 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4266; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_v_d = __pyx_t_1;
 
-  /* "mtrand.pyx":4265
+  /* "mtrand.pyx":4267
  * 
  *         d = len(pvals)
  *         parr = <ndarray>PyArray_ContiguousFromObject(pvals, NPY_DOUBLE, 1, 1)             # <<<<<<<<<<<<<<
  *         pix = <double*>PyArray_DATA(parr)
  * 
  */
-  __pyx_t_2 = PyArray_ContiguousFromObject(__pyx_v_pvals, NPY_DOUBLE, 1, 1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4265; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_ContiguousFromObject(__pyx_v_pvals, NPY_DOUBLE, 1, 1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4267; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -20359,7 +20423,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
   arrayObject_parr = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4266
+  /* "mtrand.pyx":4268
  *         d = len(pvals)
  *         parr = <ndarray>PyArray_ContiguousFromObject(pvals, NPY_DOUBLE, 1, 1)
  *         pix = <double*>PyArray_DATA(parr)             # <<<<<<<<<<<<<<
@@ -20368,7 +20432,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
  */
   __pyx_v_pix = ((double *)PyArray_DATA(arrayObject_parr));
 
-  /* "mtrand.pyx":4268
+  /* "mtrand.pyx":4270
  *         pix = <double*>PyArray_DATA(parr)
  * 
  *         if kahan_sum(pix, d-1) > (1.0 + 1e-12):             # <<<<<<<<<<<<<<
@@ -20378,34 +20442,34 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
   __pyx_t_4 = ((__pyx_f_6mtrand_kahan_sum(__pyx_v_pix, (__pyx_v_d - 1)) > (1.0 + 1e-12)) != 0);
   if (__pyx_t_4) {
 
-    /* "mtrand.pyx":4269
+    /* "mtrand.pyx":4271
  * 
  *         if kahan_sum(pix, d-1) > (1.0 + 1e-12):
  *             raise ValueError("sum(pvals[:-1]) > 1.0")             # <<<<<<<<<<<<<<
  * 
  *         shape = _shape_from_size(size, d)
  */
-    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_198), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4269; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_builtin_ValueError, ((PyObject *)__pyx_k_tuple_200), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_Raise(__pyx_t_3, 0, 0, 0);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4269; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     goto __pyx_L3;
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":4271
+  /* "mtrand.pyx":4273
  *             raise ValueError("sum(pvals[:-1]) > 1.0")
  * 
  *         shape = _shape_from_size(size, d)             # <<<<<<<<<<<<<<
  * 
  *         multin = np.zeros(shape, int)
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s___shape_from_size); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s___shape_from_size); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyInt_to_py_npy_intp(__pyx_v_d); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyInt_to_py_npy_intp(__pyx_v_d); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(2); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(__pyx_v_size);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_v_size);
@@ -20413,26 +20477,26 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
   PyTuple_SET_ITEM(__pyx_t_5, 1, __pyx_t_2);
   __Pyx_GIVEREF(__pyx_t_2);
   __pyx_t_2 = 0;
-  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
   __pyx_v_shape = __pyx_t_2;
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":4273
+  /* "mtrand.pyx":4275
  *         shape = _shape_from_size(size, d)
  * 
  *         multin = np.zeros(shape, int)             # <<<<<<<<<<<<<<
  *         mnarr = <ndarray>multin
  *         mnix = <long*>PyArray_DATA(mnarr)
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4275; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__zeros); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__zeros); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4275; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4275; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_v_shape);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_v_shape);
@@ -20440,14 +20504,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
   __Pyx_INCREF(((PyObject *)((PyObject*)(&PyInt_Type))));
   PyTuple_SET_ITEM(__pyx_t_2, 1, ((PyObject *)((PyObject*)(&PyInt_Type))));
   __Pyx_GIVEREF(((PyObject *)((PyObject*)(&PyInt_Type))));
-  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4273; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_5, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4275; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
   __pyx_v_multin = __pyx_t_3;
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4274
+  /* "mtrand.pyx":4276
  * 
  *         multin = np.zeros(shape, int)
  *         mnarr = <ndarray>multin             # <<<<<<<<<<<<<<
@@ -20457,7 +20521,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
   __Pyx_INCREF(((PyObject *)((PyArrayObject *)__pyx_v_multin)));
   arrayObject_mnarr = ((PyArrayObject *)__pyx_v_multin);
 
-  /* "mtrand.pyx":4275
+  /* "mtrand.pyx":4277
  *         multin = np.zeros(shape, int)
  *         mnarr = <ndarray>multin
  *         mnix = <long*>PyArray_DATA(mnarr)             # <<<<<<<<<<<<<<
@@ -20466,7 +20530,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
  */
   __pyx_v_mnix = ((long *)PyArray_DATA(arrayObject_mnarr));
 
-  /* "mtrand.pyx":4276
+  /* "mtrand.pyx":4278
  *         mnarr = <ndarray>multin
  *         mnix = <long*>PyArray_DATA(mnarr)
  *         i = 0             # <<<<<<<<<<<<<<
@@ -20475,7 +20539,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
  */
   __pyx_v_i = 0;
 
-  /* "mtrand.pyx":4277
+  /* "mtrand.pyx":4279
  *         mnix = <long*>PyArray_DATA(mnarr)
  *         i = 0
  *         while i < PyArray_SIZE(mnarr):             # <<<<<<<<<<<<<<
@@ -20486,7 +20550,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
     __pyx_t_4 = ((__pyx_v_i < PyArray_SIZE(arrayObject_mnarr)) != 0);
     if (!__pyx_t_4) break;
 
-    /* "mtrand.pyx":4278
+    /* "mtrand.pyx":4280
  *         i = 0
  *         while i < PyArray_SIZE(mnarr):
  *             Sum = 1.0             # <<<<<<<<<<<<<<
@@ -20495,7 +20559,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
  */
     __pyx_v_Sum = 1.0;
 
-    /* "mtrand.pyx":4279
+    /* "mtrand.pyx":4281
  *         while i < PyArray_SIZE(mnarr):
  *             Sum = 1.0
  *             dn = n             # <<<<<<<<<<<<<<
@@ -20504,7 +20568,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
  */
     __pyx_v_dn = __pyx_v_n;
 
-    /* "mtrand.pyx":4280
+    /* "mtrand.pyx":4282
  *             Sum = 1.0
  *             dn = n
  *             for j from 0 <= j < d-1:             # <<<<<<<<<<<<<<
@@ -20514,7 +20578,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
     __pyx_t_6 = (__pyx_v_d - 1);
     for (__pyx_v_j = 0; __pyx_v_j < __pyx_t_6; __pyx_v_j++) {
 
-      /* "mtrand.pyx":4281
+      /* "mtrand.pyx":4283
  *             dn = n
  *             for j from 0 <= j < d-1:
  *                 mnix[i+j] = rk_binomial(self.internal_state, dn, pix[j]/Sum)             # <<<<<<<<<<<<<<
@@ -20529,11 +20593,11 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
         #ifdef WITH_THREAD
         PyGILState_Release(__pyx_gilstate_save);
         #endif
-        {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4281; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+        {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4283; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       }
       (__pyx_v_mnix[(__pyx_v_i + __pyx_v_j)]) = rk_binomial(__pyx_v_self->internal_state, __pyx_v_dn, ((__pyx_v_pix[__pyx_v_j]) / __pyx_v_Sum));
 
-      /* "mtrand.pyx":4282
+      /* "mtrand.pyx":4284
  *             for j from 0 <= j < d-1:
  *                 mnix[i+j] = rk_binomial(self.internal_state, dn, pix[j]/Sum)
  *                 dn = dn - mnix[i+j]             # <<<<<<<<<<<<<<
@@ -20542,7 +20606,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
  */
       __pyx_v_dn = (__pyx_v_dn - (__pyx_v_mnix[(__pyx_v_i + __pyx_v_j)]));
 
-      /* "mtrand.pyx":4283
+      /* "mtrand.pyx":4285
  *                 mnix[i+j] = rk_binomial(self.internal_state, dn, pix[j]/Sum)
  *                 dn = dn - mnix[i+j]
  *                 if dn <= 0:             # <<<<<<<<<<<<<<
@@ -20552,7 +20616,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
       __pyx_t_4 = ((__pyx_v_dn <= 0) != 0);
       if (__pyx_t_4) {
 
-        /* "mtrand.pyx":4284
+        /* "mtrand.pyx":4286
  *                 dn = dn - mnix[i+j]
  *                 if dn <= 0:
  *                     break             # <<<<<<<<<<<<<<
@@ -20564,7 +20628,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
       }
       __pyx_L8:;
 
-      /* "mtrand.pyx":4285
+      /* "mtrand.pyx":4287
  *                 if dn <= 0:
  *                     break
  *                 Sum = Sum - pix[j]             # <<<<<<<<<<<<<<
@@ -20575,7 +20639,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
     }
     __pyx_L7_break:;
 
-    /* "mtrand.pyx":4286
+    /* "mtrand.pyx":4288
  *                     break
  *                 Sum = Sum - pix[j]
  *             if dn > 0:             # <<<<<<<<<<<<<<
@@ -20585,7 +20649,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
     __pyx_t_4 = ((__pyx_v_dn > 0) != 0);
     if (__pyx_t_4) {
 
-      /* "mtrand.pyx":4287
+      /* "mtrand.pyx":4289
  *                 Sum = Sum - pix[j]
  *             if dn > 0:
  *                 mnix[i+d-1] = dn             # <<<<<<<<<<<<<<
@@ -20597,7 +20661,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
     }
     __pyx_L9:;
 
-    /* "mtrand.pyx":4289
+    /* "mtrand.pyx":4291
  *                 mnix[i+d-1] = dn
  * 
  *             i = i + d             # <<<<<<<<<<<<<<
@@ -20607,7 +20671,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_98multinomial(struct __pyx_obj_6
     __pyx_v_i = (__pyx_v_i + __pyx_v_d);
   }
 
-  /* "mtrand.pyx":4291
+  /* "mtrand.pyx":4293
  *             i = i + d
  * 
  *         return multin             # <<<<<<<<<<<<<<
@@ -20653,7 +20717,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_101dirichlet(PyObject *__pyx_v_s
     static PyObject **__pyx_pyargnames[] = {&__pyx_n_s__alpha,&__pyx_n_s__size,0};
     PyObject* values[2] = {0,0};
 
-    /* "mtrand.pyx":4293
+    /* "mtrand.pyx":4295
  *         return multin
  * 
  *     def dirichlet(self, object alpha, size=None):             # <<<<<<<<<<<<<<
@@ -20682,7 +20746,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_101dirichlet(PyObject *__pyx_v_s
         }
       }
       if (unlikely(kw_args > 0)) {
-        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "dirichlet") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4293; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+        if (unlikely(__Pyx_ParseOptionalKeywords(__pyx_kwds, __pyx_pyargnames, 0, values, pos_args, "dirichlet") < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4295; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
       }
     } else {
       switch (PyTuple_GET_SIZE(__pyx_args)) {
@@ -20697,7 +20761,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_101dirichlet(PyObject *__pyx_v_s
   }
   goto __pyx_L4_argument_unpacking_done;
   __pyx_L5_argtuple_error:;
-  __Pyx_RaiseArgtupleInvalid("dirichlet", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4293; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
+  __Pyx_RaiseArgtupleInvalid("dirichlet", 0, 1, 2, PyTuple_GET_SIZE(__pyx_args)); {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4295; __pyx_clineno = __LINE__; goto __pyx_L3_error;}
   __pyx_L3_error:;
   __Pyx_AddTraceback("mtrand.RandomState.dirichlet", __pyx_clineno, __pyx_lineno, __pyx_filename);
   __Pyx_RefNannyFinishContext();
@@ -20734,24 +20798,24 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("dirichlet", 0);
 
-  /* "mtrand.pyx":4380
+  /* "mtrand.pyx":4382
  *         cdef double     acc, invacc
  * 
  *         k           = len(alpha)             # <<<<<<<<<<<<<<
  *         alpha_arr   = <ndarray>PyArray_ContiguousFromObject(alpha, NPY_DOUBLE, 1, 1)
  *         alpha_data  = <double*>PyArray_DATA(alpha_arr)
  */
-  __pyx_t_1 = PyObject_Length(__pyx_v_alpha); if (unlikely(__pyx_t_1 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4380; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Length(__pyx_v_alpha); if (unlikely(__pyx_t_1 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4382; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_v_k = __pyx_t_1;
 
-  /* "mtrand.pyx":4381
+  /* "mtrand.pyx":4383
  * 
  *         k           = len(alpha)
  *         alpha_arr   = <ndarray>PyArray_ContiguousFromObject(alpha, NPY_DOUBLE, 1, 1)             # <<<<<<<<<<<<<<
  *         alpha_data  = <double*>PyArray_DATA(alpha_arr)
  * 
  */
-  __pyx_t_2 = PyArray_ContiguousFromObject(__pyx_v_alpha, NPY_DOUBLE, 1, 1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4381; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyArray_ContiguousFromObject(__pyx_v_alpha, NPY_DOUBLE, 1, 1); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4383; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __pyx_t_3 = __pyx_t_2;
   __Pyx_INCREF(__pyx_t_3);
@@ -20759,7 +20823,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
   __pyx_v_alpha_arr = ((PyArrayObject *)__pyx_t_3);
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4382
+  /* "mtrand.pyx":4384
  *         k           = len(alpha)
  *         alpha_arr   = <ndarray>PyArray_ContiguousFromObject(alpha, NPY_DOUBLE, 1, 1)
  *         alpha_data  = <double*>PyArray_DATA(alpha_arr)             # <<<<<<<<<<<<<<
@@ -20768,18 +20832,18 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
  */
   __pyx_v_alpha_data = ((double *)PyArray_DATA(__pyx_v_alpha_arr));
 
-  /* "mtrand.pyx":4384
+  /* "mtrand.pyx":4386
  *         alpha_data  = <double*>PyArray_DATA(alpha_arr)
  * 
  *         shape = _shape_from_size(size, k)             # <<<<<<<<<<<<<<
  * 
  *         diric   = np.zeros(shape, np.float64)
  */
-  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s___shape_from_size); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4384; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s___shape_from_size); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
-  __pyx_t_2 = __Pyx_PyInt_to_py_npy_intp(__pyx_v_k); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4384; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyInt_to_py_npy_intp(__pyx_v_k); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4384; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyTuple_New(2); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_INCREF(__pyx_v_size);
   PyTuple_SET_ITEM(__pyx_t_4, 0, __pyx_v_size);
@@ -20787,31 +20851,31 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
   PyTuple_SET_ITEM(__pyx_t_4, 1, __pyx_t_2);
   __Pyx_GIVEREF(__pyx_t_2);
   __pyx_t_2 = 0;
-  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4384; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyObject_Call(__pyx_t_3, ((PyObject *)__pyx_t_4), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_4)); __pyx_t_4 = 0;
   __pyx_v_shape = __pyx_t_2;
   __pyx_t_2 = 0;
 
-  /* "mtrand.pyx":4386
+  /* "mtrand.pyx":4388
  *         shape = _shape_from_size(size, k)
  * 
  *         diric   = np.zeros(shape, np.float64)             # <<<<<<<<<<<<<<
  *         val_arr = <ndarray>diric
  *         val_data= <double*>PyArray_DATA(val_arr)
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4388; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__zeros); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__zeros); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4388; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4388; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__float64); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__float64); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4388; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = PyTuple_New(2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4388; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_INCREF(__pyx_v_shape);
   PyTuple_SET_ITEM(__pyx_t_2, 0, __pyx_v_shape);
@@ -20819,14 +20883,14 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
   PyTuple_SET_ITEM(__pyx_t_2, 1, __pyx_t_3);
   __Pyx_GIVEREF(__pyx_t_3);
   __pyx_t_3 = 0;
-  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4386; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_t_2), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4388; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_2)); __pyx_t_2 = 0;
   __pyx_v_diric = __pyx_t_3;
   __pyx_t_3 = 0;
 
-  /* "mtrand.pyx":4387
+  /* "mtrand.pyx":4389
  * 
  *         diric   = np.zeros(shape, np.float64)
  *         val_arr = <ndarray>diric             # <<<<<<<<<<<<<<
@@ -20836,7 +20900,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
   __Pyx_INCREF(((PyObject *)((PyArrayObject *)__pyx_v_diric)));
   __pyx_v_val_arr = ((PyArrayObject *)__pyx_v_diric);
 
-  /* "mtrand.pyx":4388
+  /* "mtrand.pyx":4390
  *         diric   = np.zeros(shape, np.float64)
  *         val_arr = <ndarray>diric
  *         val_data= <double*>PyArray_DATA(val_arr)             # <<<<<<<<<<<<<<
@@ -20845,7 +20909,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
  */
   __pyx_v_val_data = ((double *)PyArray_DATA(__pyx_v_val_arr));
 
-  /* "mtrand.pyx":4390
+  /* "mtrand.pyx":4392
  *         val_data= <double*>PyArray_DATA(val_arr)
  * 
  *         i = 0             # <<<<<<<<<<<<<<
@@ -20854,7 +20918,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
  */
   __pyx_v_i = 0;
 
-  /* "mtrand.pyx":4391
+  /* "mtrand.pyx":4393
  * 
  *         i = 0
  *         totsize = PyArray_SIZE(val_arr)             # <<<<<<<<<<<<<<
@@ -20863,7 +20927,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
  */
   __pyx_v_totsize = PyArray_SIZE(__pyx_v_val_arr);
 
-  /* "mtrand.pyx":4392
+  /* "mtrand.pyx":4394
  *         i = 0
  *         totsize = PyArray_SIZE(val_arr)
  *         while i < totsize:             # <<<<<<<<<<<<<<
@@ -20874,7 +20938,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
     __pyx_t_5 = ((__pyx_v_i < __pyx_v_totsize) != 0);
     if (!__pyx_t_5) break;
 
-    /* "mtrand.pyx":4393
+    /* "mtrand.pyx":4395
  *         totsize = PyArray_SIZE(val_arr)
  *         while i < totsize:
  *             acc = 0.0             # <<<<<<<<<<<<<<
@@ -20883,7 +20947,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
  */
     __pyx_v_acc = 0.0;
 
-    /* "mtrand.pyx":4394
+    /* "mtrand.pyx":4396
  *         while i < totsize:
  *             acc = 0.0
  *             for j from 0 <= j < k:             # <<<<<<<<<<<<<<
@@ -20893,7 +20957,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
     __pyx_t_6 = __pyx_v_k;
     for (__pyx_v_j = 0; __pyx_v_j < __pyx_t_6; __pyx_v_j++) {
 
-      /* "mtrand.pyx":4395
+      /* "mtrand.pyx":4397
  *             acc = 0.0
  *             for j from 0 <= j < k:
  *                 val_data[i+j]   = rk_standard_gamma(self.internal_state, alpha_data[j])             # <<<<<<<<<<<<<<
@@ -20902,7 +20966,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
  */
       (__pyx_v_val_data[(__pyx_v_i + __pyx_v_j)]) = rk_standard_gamma(__pyx_v_self->internal_state, (__pyx_v_alpha_data[__pyx_v_j]));
 
-      /* "mtrand.pyx":4396
+      /* "mtrand.pyx":4398
  *             for j from 0 <= j < k:
  *                 val_data[i+j]   = rk_standard_gamma(self.internal_state, alpha_data[j])
  *                 acc             = acc + val_data[i+j]             # <<<<<<<<<<<<<<
@@ -20912,7 +20976,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
       __pyx_v_acc = (__pyx_v_acc + (__pyx_v_val_data[(__pyx_v_i + __pyx_v_j)]));
     }
 
-    /* "mtrand.pyx":4397
+    /* "mtrand.pyx":4399
  *                 val_data[i+j]   = rk_standard_gamma(self.internal_state, alpha_data[j])
  *                 acc             = acc + val_data[i+j]
  *             invacc  = 1/acc             # <<<<<<<<<<<<<<
@@ -20927,11 +20991,11 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
       #ifdef WITH_THREAD
       PyGILState_Release(__pyx_gilstate_save);
       #endif
-      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4397; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4399; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     }
     __pyx_v_invacc = (1.0 / __pyx_v_acc);
 
-    /* "mtrand.pyx":4398
+    /* "mtrand.pyx":4400
  *                 acc             = acc + val_data[i+j]
  *             invacc  = 1/acc
  *             for j from 0 <= j < k:             # <<<<<<<<<<<<<<
@@ -20941,7 +21005,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
     __pyx_t_6 = __pyx_v_k;
     for (__pyx_v_j = 0; __pyx_v_j < __pyx_t_6; __pyx_v_j++) {
 
-      /* "mtrand.pyx":4399
+      /* "mtrand.pyx":4401
  *             invacc  = 1/acc
  *             for j from 0 <= j < k:
  *                 val_data[i+j]   = val_data[i+j] * invacc             # <<<<<<<<<<<<<<
@@ -20951,7 +21015,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
       (__pyx_v_val_data[(__pyx_v_i + __pyx_v_j)]) = ((__pyx_v_val_data[(__pyx_v_i + __pyx_v_j)]) * __pyx_v_invacc);
     }
 
-    /* "mtrand.pyx":4400
+    /* "mtrand.pyx":4402
  *             for j from 0 <= j < k:
  *                 val_data[i+j]   = val_data[i+j] * invacc
  *             i = i + k             # <<<<<<<<<<<<<<
@@ -20961,7 +21025,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_100dirichlet(struct __pyx_obj_6m
     __pyx_v_i = (__pyx_v_i + __pyx_v_k);
   }
 
-  /* "mtrand.pyx":4402
+  /* "mtrand.pyx":4404
  *             i = i + k
  * 
  *         return diric             # <<<<<<<<<<<<<<
@@ -21003,7 +21067,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_103shuffle(PyObject *__pyx_v_sel
   return __pyx_r;
 }
 
-/* "mtrand.pyx":4405
+/* "mtrand.pyx":4407
  * 
  *     # Shuffling and permutations:
  *     def shuffle(self, object x):             # <<<<<<<<<<<<<<
@@ -21030,49 +21094,49 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("shuffle", 0);
 
-  /* "mtrand.pyx":4440
+  /* "mtrand.pyx":4442
  *         cdef npy_intp i, j
  * 
  *         i = len(x) - 1             # <<<<<<<<<<<<<<
  * 
  *         # Logic adapted from random.shuffle()
  */
-  __pyx_t_1 = PyObject_Length(__pyx_v_x); if (unlikely(__pyx_t_1 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4440; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Length(__pyx_v_x); if (unlikely(__pyx_t_1 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4442; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_v_i = (__pyx_t_1 - 1);
 
-  /* "mtrand.pyx":4443
+  /* "mtrand.pyx":4445
  * 
  *         # Logic adapted from random.shuffle()
  *         if isinstance(x, np.ndarray) and \             # <<<<<<<<<<<<<<
  *            (x.ndim > 1 or x.dtype.fields is not None):
  *             # For a multi-dimensional ndarray, indexing returns a view onto
  */
-  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4443; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4445; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__ndarray); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4443; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__ndarray); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4445; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_3);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
-  __pyx_t_4 = PyObject_IsInstance(__pyx_v_x, __pyx_t_3); if (unlikely(__pyx_t_4 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4443; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_IsInstance(__pyx_v_x, __pyx_t_3); if (unlikely(__pyx_t_4 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4445; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
   if (__pyx_t_4) {
 
-    /* "mtrand.pyx":4444
+    /* "mtrand.pyx":4446
  *         # Logic adapted from random.shuffle()
  *         if isinstance(x, np.ndarray) and \
  *            (x.ndim > 1 or x.dtype.fields is not None):             # <<<<<<<<<<<<<<
  *             # For a multi-dimensional ndarray, indexing returns a view onto
  *             # each row. So we can't just use ordinary assignment to swap the
  */
-    __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_v_x, __pyx_n_s__ndim); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4444; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_v_x, __pyx_n_s__ndim); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4446; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
-    __pyx_t_2 = PyObject_RichCompare(__pyx_t_3, __pyx_int_1, Py_GT); __Pyx_XGOTREF(__pyx_t_2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4444; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = PyObject_RichCompare(__pyx_t_3, __pyx_int_1, Py_GT); __Pyx_XGOTREF(__pyx_t_2); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4446; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    __pyx_t_5 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_5 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4444; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = __Pyx_PyObject_IsTrue(__pyx_t_2); if (unlikely(__pyx_t_5 < 0)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4446; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
     if (!__pyx_t_5) {
-      __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_v_x, __pyx_n_s__dtype); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4444; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_v_x, __pyx_n_s__dtype); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4446; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_2);
-      __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__fields); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4444; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_3 = __Pyx_PyObject_GetAttrStr(__pyx_t_2, __pyx_n_s__fields); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4446; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_3);
       __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
       __pyx_t_6 = (__pyx_t_3 != Py_None);
@@ -21087,33 +21151,33 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
   }
   if (__pyx_t_5) {
 
-    /* "mtrand.pyx":4448
+    /* "mtrand.pyx":4450
  *             # each row. So we can't just use ordinary assignment to swap the
  *             # rows; we need a bounce buffer.
  *             buf = np.empty_like(x[0])             # <<<<<<<<<<<<<<
  *             while i > 0:
  *                 j = rk_interval(i, self.internal_state)
  */
-    __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4448; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4450; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
-    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__empty_like); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4448; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_3, __pyx_n_s__empty_like); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4450; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-    __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4448; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, 0, sizeof(long), PyInt_FromLong, 0, 0, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4450; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
-    __pyx_t_8 = PyTuple_New(1); if (unlikely(!__pyx_t_8)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4448; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_8 = PyTuple_New(1); if (unlikely(!__pyx_t_8)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4450; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_8);
     PyTuple_SET_ITEM(__pyx_t_8, 0, __pyx_t_3);
     __Pyx_GIVEREF(__pyx_t_3);
     __pyx_t_3 = 0;
-    __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_8), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4448; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_3 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_8), NULL); if (unlikely(!__pyx_t_3)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4450; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_3);
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
     __Pyx_DECREF(((PyObject *)__pyx_t_8)); __pyx_t_8 = 0;
     __pyx_v_buf = __pyx_t_3;
     __pyx_t_3 = 0;
 
-    /* "mtrand.pyx":4449
+    /* "mtrand.pyx":4451
  *             # rows; we need a bounce buffer.
  *             buf = np.empty_like(x[0])
  *             while i > 0:             # <<<<<<<<<<<<<<
@@ -21124,7 +21188,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
       __pyx_t_5 = ((__pyx_v_i > 0) != 0);
       if (!__pyx_t_5) break;
 
-      /* "mtrand.pyx":4450
+      /* "mtrand.pyx":4452
  *             buf = np.empty_like(x[0])
  *             while i > 0:
  *                 j = rk_interval(i, self.internal_state)             # <<<<<<<<<<<<<<
@@ -21133,40 +21197,40 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
  */
       __pyx_v_j = rk_interval(__pyx_v_i, __pyx_v_self->internal_state);
 
-      /* "mtrand.pyx":4451
+      /* "mtrand.pyx":4453
  *             while i > 0:
  *                 j = rk_interval(i, self.internal_state)
  *                 buf[...] = x[j]             # <<<<<<<<<<<<<<
  *                 x[j] = x[i]
  *                 x[i] = buf
  */
-      __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_j, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4451; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_j, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4453; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_3);
-      if (PyObject_SetItem(__pyx_v_buf, Py_Ellipsis, __pyx_t_3) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4451; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      if (PyObject_SetItem(__pyx_v_buf, Py_Ellipsis, __pyx_t_3) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4453; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
 
-      /* "mtrand.pyx":4452
+      /* "mtrand.pyx":4454
  *                 j = rk_interval(i, self.internal_state)
  *                 buf[...] = x[j]
  *                 x[j] = x[i]             # <<<<<<<<<<<<<<
  *                 x[i] = buf
  *                 i = i - 1
  */
-      __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_i, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4452; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_i, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4454; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_3);
-      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_j, __pyx_t_3, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4452; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_j, __pyx_t_3, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4454; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
 
-      /* "mtrand.pyx":4453
+      /* "mtrand.pyx":4455
  *                 buf[...] = x[j]
  *                 x[j] = x[i]
  *                 x[i] = buf             # <<<<<<<<<<<<<<
  *                 i = i - 1
  *         else:
  */
-      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_i, __pyx_v_buf, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4453; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_i, __pyx_v_buf, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4455; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
 
-      /* "mtrand.pyx":4454
+      /* "mtrand.pyx":4456
  *                 x[j] = x[i]
  *                 x[i] = buf
  *                 i = i - 1             # <<<<<<<<<<<<<<
@@ -21179,7 +21243,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
   }
   /*else*/ {
 
-    /* "mtrand.pyx":4459
+    /* "mtrand.pyx":4461
  *             # sequence types, indexing returns a real object that's
  *             # independent of the array contents, so we can just swap directly.
  *             while i > 0:             # <<<<<<<<<<<<<<
@@ -21190,7 +21254,7 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
       __pyx_t_5 = ((__pyx_v_i > 0) != 0);
       if (!__pyx_t_5) break;
 
-      /* "mtrand.pyx":4460
+      /* "mtrand.pyx":4462
  *             # independent of the array contents, so we can just swap directly.
  *             while i > 0:
  *                 j = rk_interval(i, self.internal_state)             # <<<<<<<<<<<<<<
@@ -21199,23 +21263,23 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_102shuffle(struct __pyx_obj_6mtr
  */
       __pyx_v_j = rk_interval(__pyx_v_i, __pyx_v_self->internal_state);
 
-      /* "mtrand.pyx":4461
+      /* "mtrand.pyx":4463
  *             while i > 0:
  *                 j = rk_interval(i, self.internal_state)
  *                 x[i], x[j] = x[j], x[i]             # <<<<<<<<<<<<<<
  *                 i = i - 1
  * 
  */
-      __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_j, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4461; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_3 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_j, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_3) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4463; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_3);
-      __pyx_t_8 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_i, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_8) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4461; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      __pyx_t_8 = __Pyx_GetItemInt(__pyx_v_x, __pyx_v_i, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1); if (!__pyx_t_8) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4463; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_GOTREF(__pyx_t_8);
-      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_i, __pyx_t_3, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4461; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_i, __pyx_t_3, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4463; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_DECREF(__pyx_t_3); __pyx_t_3 = 0;
-      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_j, __pyx_t_8, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4461; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+      if (__Pyx_SetItemInt(__pyx_v_x, __pyx_v_j, __pyx_t_8, sizeof(npy_intp), __Pyx_PyInt_to_py_npy_intp, 0, 1, 1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4463; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
       __Pyx_DECREF(__pyx_t_8); __pyx_t_8 = 0;
 
-      /* "mtrand.pyx":4462
+      /* "mtrand.pyx":4464
  *                 j = rk_interval(i, self.internal_state)
  *                 x[i], x[j] = x[j], x[i]
  *                 i = i - 1             # <<<<<<<<<<<<<<
@@ -21254,7 +21318,7 @@ static PyObject *__pyx_pw_6mtrand_11RandomState_105permutation(PyObject *__pyx_v
   return __pyx_r;
 }
 
-/* "mtrand.pyx":4464
+/* "mtrand.pyx":4466
  *                 i = i - 1
  * 
  *     def permutation(self, object x):             # <<<<<<<<<<<<<<
@@ -21276,19 +21340,19 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_104permutation(struct __pyx_obj_
   int __pyx_clineno = 0;
   __Pyx_RefNannySetupContext("permutation", 0);
 
-  /* "mtrand.pyx":4500
+  /* "mtrand.pyx":4502
  * 
  *         """
  *         if isinstance(x, (int, long, np.integer)):             # <<<<<<<<<<<<<<
  *             arr = np.arange(x)
  *         else:
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4500; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4502; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__integer); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4500; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__integer); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4502; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_1 = PyTuple_New(3); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4500; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyTuple_New(3); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4502; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_INCREF(((PyObject *)((PyObject*)(&PyInt_Type))));
   PyTuple_SET_ITEM(__pyx_t_1, 0, ((PyObject *)((PyObject*)(&PyInt_Type))));
@@ -21299,29 +21363,29 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_104permutation(struct __pyx_obj_
   PyTuple_SET_ITEM(__pyx_t_1, 2, __pyx_t_2);
   __Pyx_GIVEREF(__pyx_t_2);
   __pyx_t_2 = 0;
-  __pyx_t_3 = PyObject_IsInstance(__pyx_v_x, ((PyObject *)__pyx_t_1)); if (unlikely(__pyx_t_3 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4500; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_3 = PyObject_IsInstance(__pyx_v_x, ((PyObject *)__pyx_t_1)); if (unlikely(__pyx_t_3 == -1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4502; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
   __pyx_t_4 = (__pyx_t_3 != 0);
   if (__pyx_t_4) {
 
-    /* "mtrand.pyx":4501
+    /* "mtrand.pyx":4503
  *         """
  *         if isinstance(x, (int, long, np.integer)):
  *             arr = np.arange(x)             # <<<<<<<<<<<<<<
  *         else:
  *             arr = np.array(x)
  */
-    __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4501; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
-    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__arange); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4501; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__arange); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-    __pyx_t_1 = PyTuple_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4501; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = PyTuple_New(1); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __Pyx_INCREF(__pyx_v_x);
     PyTuple_SET_ITEM(__pyx_t_1, 0, __pyx_v_x);
     __Pyx_GIVEREF(__pyx_v_x);
-    __pyx_t_5 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_1), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4501; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_1), NULL); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
     __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
@@ -21331,24 +21395,24 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_104permutation(struct __pyx_obj_
   }
   /*else*/ {
 
-    /* "mtrand.pyx":4503
+    /* "mtrand.pyx":4505
  *             arr = np.arange(x)
  *         else:
  *             arr = np.array(x)             # <<<<<<<<<<<<<<
  *         self.shuffle(arr)
  *         return arr
  */
-    __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = __Pyx_GetModuleGlobalName(__pyx_n_s__np); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4505; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
-    __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__array); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_5, __pyx_n_s__array); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4505; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_1);
     __Pyx_DECREF(__pyx_t_5); __pyx_t_5 = 0;
-    __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4505; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_5);
     __Pyx_INCREF(__pyx_v_x);
     PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_v_x);
     __Pyx_GIVEREF(__pyx_v_x);
-    __pyx_t_2 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4503; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+    __pyx_t_2 = PyObject_Call(__pyx_t_1, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4505; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
     __Pyx_GOTREF(__pyx_t_2);
     __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
     __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
@@ -21357,27 +21421,27 @@ static PyObject *__pyx_pf_6mtrand_11RandomState_104permutation(struct __pyx_obj_
   }
   __pyx_L3:;
 
-  /* "mtrand.pyx":4504
+  /* "mtrand.pyx":4506
  *         else:
  *             arr = np.array(x)
  *         self.shuffle(arr)             # <<<<<<<<<<<<<<
  *         return arr
  * 
  */
-  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__shuffle); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4504; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_2 = __Pyx_PyObject_GetAttrStr(((PyObject *)__pyx_v_self), __pyx_n_s__shuffle); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4506; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
-  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4504; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_5 = PyTuple_New(1); if (unlikely(!__pyx_t_5)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4506; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_5);
   __Pyx_INCREF(__pyx_v_arr);
   PyTuple_SET_ITEM(__pyx_t_5, 0, __pyx_v_arr);
   __Pyx_GIVEREF(__pyx_v_arr);
-  __pyx_t_1 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4504; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_t_5), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4506; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __Pyx_DECREF(((PyObject *)__pyx_t_5)); __pyx_t_5 = 0;
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4505
+  /* "mtrand.pyx":4507
  *             arr = np.array(x)
  *         self.shuffle(arr)
  *         return arr             # <<<<<<<<<<<<<<
@@ -21533,9 +21597,6 @@ static PyTypeObject __pyx_type_6mtrand_RandomState = {
   #if PY_VERSION_HEX >= 0x02060000
   0, /*tp_version_tag*/
   #endif
-  #if PY_VERSION_HEX >= 0x030400a1 && defined(Py_TPFLAGS_HAVE_FINALIZE)
-  0, /*tp_finalize*/
-  #endif
 };
 
 static PyMethodDef __pyx_methods[] = {
@@ -21574,33 +21635,32 @@ static __Pyx_StringTabEntry __pyx_string_tab[] = {
   {&__pyx_kp_s_135, __pyx_k_135, sizeof(__pyx_k_135), 0, 0, 1, 0},
   {&__pyx_kp_s_137, __pyx_k_137, sizeof(__pyx_k_137), 0, 0, 1, 0},
   {&__pyx_kp_s_139, __pyx_k_139, sizeof(__pyx_k_139), 0, 0, 1, 0},
-  {&__pyx_kp_s_144, __pyx_k_144, sizeof(__pyx_k_144), 0, 0, 1, 0},
+  {&__pyx_kp_s_141, __pyx_k_141, sizeof(__pyx_k_141), 0, 0, 1, 0},
+  {&__pyx_kp_s_146, __pyx_k_146, sizeof(__pyx_k_146), 0, 0, 1, 0},
   {&__pyx_kp_s_15, __pyx_k_15, sizeof(__pyx_k_15), 0, 0, 1, 0},
-  {&__pyx_kp_s_152, __pyx_k_152, sizeof(__pyx_k_152), 0, 0, 1, 0},
   {&__pyx_kp_s_154, __pyx_k_154, sizeof(__pyx_k_154), 0, 0, 1, 0},
-  {&__pyx_kp_s_157, __pyx_k_157, sizeof(__pyx_k_157), 0, 0, 1, 0},
+  {&__pyx_kp_s_156, __pyx_k_156, sizeof(__pyx_k_156), 0, 0, 1, 0},
   {&__pyx_kp_s_159, __pyx_k_159, sizeof(__pyx_k_159), 0, 0, 1, 0},
-  {&__pyx_kp_s_162, __pyx_k_162, sizeof(__pyx_k_162), 0, 0, 1, 0},
+  {&__pyx_kp_s_161, __pyx_k_161, sizeof(__pyx_k_161), 0, 0, 1, 0},
   {&__pyx_kp_s_164, __pyx_k_164, sizeof(__pyx_k_164), 0, 0, 1, 0},
-  {&__pyx_kp_s_168, __pyx_k_168, sizeof(__pyx_k_168), 0, 0, 1, 0},
+  {&__pyx_kp_s_166, __pyx_k_166, sizeof(__pyx_k_166), 0, 0, 1, 0},
   {&__pyx_kp_s_170, __pyx_k_170, sizeof(__pyx_k_170), 0, 0, 1, 0},
   {&__pyx_kp_s_172, __pyx_k_172, sizeof(__pyx_k_172), 0, 0, 1, 0},
   {&__pyx_kp_s_174, __pyx_k_174, sizeof(__pyx_k_174), 0, 0, 1, 0},
+  {&__pyx_kp_s_176, __pyx_k_176, sizeof(__pyx_k_176), 0, 0, 1, 0},
   {&__pyx_kp_s_18, __pyx_k_18, sizeof(__pyx_k_18), 0, 0, 1, 0},
-  {&__pyx_kp_s_180, __pyx_k_180, sizeof(__pyx_k_180), 0, 0, 1, 0},
   {&__pyx_kp_s_182, __pyx_k_182, sizeof(__pyx_k_182), 0, 0, 1, 0},
-  {&__pyx_n_s_186, __pyx_k_186, sizeof(__pyx_k_186), 0, 0, 1, 1},
-  {&__pyx_kp_s_187, __pyx_k_187, sizeof(__pyx_k_187), 0, 0, 1, 0},
+  {&__pyx_kp_s_184, __pyx_k_184, sizeof(__pyx_k_184), 0, 0, 1, 0},
+  {&__pyx_n_s_188, __pyx_k_188, sizeof(__pyx_k_188), 0, 0, 1, 1},
   {&__pyx_kp_s_189, __pyx_k_189, sizeof(__pyx_k_189), 0, 0, 1, 0},
   {&__pyx_kp_s_191, __pyx_k_191, sizeof(__pyx_k_191), 0, 0, 1, 0},
-  {&__pyx_kp_s_194, __pyx_k_194, sizeof(__pyx_k_194), 0, 0, 1, 0},
-  {&__pyx_kp_s_197, __pyx_k_197, sizeof(__pyx_k_197), 0, 0, 1, 0},
+  {&__pyx_kp_s_193, __pyx_k_193, sizeof(__pyx_k_193), 0, 0, 1, 0},
+  {&__pyx_kp_s_196, __pyx_k_196, sizeof(__pyx_k_196), 0, 0, 1, 0},
+  {&__pyx_kp_s_199, __pyx_k_199, sizeof(__pyx_k_199), 0, 0, 1, 0},
   {&__pyx_kp_s_20, __pyx_k_20, sizeof(__pyx_k_20), 0, 0, 1, 0},
-  {&__pyx_kp_s_201, __pyx_k_201, sizeof(__pyx_k_201), 0, 0, 1, 0},
-  {&__pyx_n_s_204, __pyx_k_204, sizeof(__pyx_k_204), 0, 0, 1, 1},
-  {&__pyx_n_s_205, __pyx_k_205, sizeof(__pyx_k_205), 0, 0, 1, 1},
-  {&__pyx_kp_u_206, __pyx_k_206, sizeof(__pyx_k_206), 0, 1, 0, 0},
-  {&__pyx_kp_u_207, __pyx_k_207, sizeof(__pyx_k_207), 0, 1, 0, 0},
+  {&__pyx_kp_s_203, __pyx_k_203, sizeof(__pyx_k_203), 0, 0, 1, 0},
+  {&__pyx_n_s_206, __pyx_k_206, sizeof(__pyx_k_206), 0, 0, 1, 1},
+  {&__pyx_n_s_207, __pyx_k_207, sizeof(__pyx_k_207), 0, 0, 1, 1},
   {&__pyx_kp_u_208, __pyx_k_208, sizeof(__pyx_k_208), 0, 1, 0, 0},
   {&__pyx_kp_u_209, __pyx_k_209, sizeof(__pyx_k_209), 0, 1, 0, 0},
   {&__pyx_kp_u_210, __pyx_k_210, sizeof(__pyx_k_210), 0, 1, 0, 0},
@@ -21689,6 +21749,8 @@ static __Pyx_StringTabEntry __pyx_string_tab[] = {
   {&__pyx_kp_u_289, __pyx_k_289, sizeof(__pyx_k_289), 0, 1, 0, 0},
   {&__pyx_kp_u_290, __pyx_k_290, sizeof(__pyx_k_290), 0, 1, 0, 0},
   {&__pyx_kp_u_291, __pyx_k_291, sizeof(__pyx_k_291), 0, 1, 0, 0},
+  {&__pyx_kp_u_292, __pyx_k_292, sizeof(__pyx_k_292), 0, 1, 0, 0},
+  {&__pyx_kp_u_293, __pyx_k_293, sizeof(__pyx_k_293), 0, 1, 0, 0},
   {&__pyx_kp_s_30, __pyx_k_30, sizeof(__pyx_k_30), 0, 0, 1, 0},
   {&__pyx_kp_s_32, __pyx_k_32, sizeof(__pyx_k_32), 0, 0, 1, 0},
   {&__pyx_kp_s_34, __pyx_k_34, sizeof(__pyx_k_34), 0, 0, 1, 0},
@@ -21762,6 +21824,7 @@ static __Pyx_StringTabEntry __pyx_string_tab[] = {
   {&__pyx_n_s__int, __pyx_k__int, sizeof(__pyx_k__int), 0, 0, 1, 1},
   {&__pyx_n_s__integer, __pyx_k__integer, sizeof(__pyx_k__integer), 0, 0, 1, 1},
   {&__pyx_n_s__intp, __pyx_k__intp, sizeof(__pyx_k__intp), 0, 0, 1, 1},
+  {&__pyx_n_s__isnan, __pyx_k__isnan, sizeof(__pyx_k__isnan), 0, 0, 1, 1},
   {&__pyx_n_s__item, __pyx_k__item, sizeof(__pyx_k__item), 0, 0, 1, 1},
   {&__pyx_n_s__kappa, __pyx_k__kappa, sizeof(__pyx_k__kappa), 0, 0, 1, 1},
   {&__pyx_n_s__l, __pyx_k__l, sizeof(__pyx_k__l), 0, 0, 1, 1},
@@ -21853,7 +21916,7 @@ static __Pyx_StringTabEntry __pyx_string_tab[] = {
 static int __Pyx_InitCachedBuiltins(void) {
   __pyx_builtin_ValueError = __Pyx_GetBuiltinName(__pyx_n_s__ValueError); if (!__pyx_builtin_ValueError) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 187; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_builtin_TypeError = __Pyx_GetBuiltinName(__pyx_n_s__TypeError); if (!__pyx_builtin_TypeError) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __pyx_builtin_RuntimeWarning = __Pyx_GetBuiltinName(__pyx_n_s__RuntimeWarning); if (!__pyx_builtin_RuntimeWarning) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4198; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_builtin_RuntimeWarning = __Pyx_GetBuiltinName(__pyx_n_s__RuntimeWarning); if (!__pyx_builtin_RuntimeWarning) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   return 0;
   __pyx_L1_error:;
   return -1;
@@ -22813,422 +22876,433 @@ static int __Pyx_InitCachedConstants(void) {
  *                 raise ValueError("p < 0")
  *             elif fp > 1:
  *                 raise ValueError("p > 1")             # <<<<<<<<<<<<<<
- *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
- * 
+ *             elif np.isnan(fp):
+ *                 raise ValueError("p is nan")
  */
   __pyx_k_tuple_140 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_140)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_k_tuple_140);
   __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_140));
 
-  /* "mtrand.pyx":3521
+  /* "mtrand.pyx":3515
+ *                 raise ValueError("p > 1")
+ *             elif np.isnan(fp):
+ *                 raise ValueError("p is nan")             # <<<<<<<<<<<<<<
+ *             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
+ * 
+ */
+  __pyx_k_tuple_142 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_141)); if (unlikely(!__pyx_k_tuple_142)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_142);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_142));
+
+  /* "mtrand.pyx":3523
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(n, 0)):
  *             raise ValueError("n < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")
  */
-  __pyx_k_tuple_141 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_135)); if (unlikely(!__pyx_k_tuple_141)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_141);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_141));
+  __pyx_k_tuple_143 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_135)); if (unlikely(!__pyx_k_tuple_143)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_143);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_143));
 
-  /* "mtrand.pyx":3523
+  /* "mtrand.pyx":3525
  *             raise ValueError("n < 0")
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")
  */
-  __pyx_k_tuple_142 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_137)); if (unlikely(!__pyx_k_tuple_142)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_142);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_142));
+  __pyx_k_tuple_144 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_137)); if (unlikely(!__pyx_k_tuple_144)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_144);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_144));
 
-  /* "mtrand.pyx":3525
+  /* "mtrand.pyx":3527
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")             # <<<<<<<<<<<<<<
  *         return discnp_array(self.internal_state, rk_binomial, size, on, op)
  * 
  */
-  __pyx_k_tuple_143 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_143)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_143);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_143));
+  __pyx_k_tuple_145 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_145)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_145);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_145));
 
-  /* "mtrand.pyx":3602
+  /* "mtrand.pyx":3604
  *         if not PyErr_Occurred():
  *             if fn <= 0:
  *                 raise ValueError("n <= 0")             # <<<<<<<<<<<<<<
  *             if fp < 0:
  *                 raise ValueError("p < 0")
  */
-  __pyx_k_tuple_145 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_144)); if (unlikely(!__pyx_k_tuple_145)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3602; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_145);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_145));
+  __pyx_k_tuple_147 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_146)); if (unlikely(!__pyx_k_tuple_147)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3604; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_147);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_147));
 
-  /* "mtrand.pyx":3604
+  /* "mtrand.pyx":3606
  *                 raise ValueError("n <= 0")
  *             if fp < 0:
  *                 raise ValueError("p < 0")             # <<<<<<<<<<<<<<
  *             elif fp > 1:
  *                 raise ValueError("p > 1")
  */
-  __pyx_k_tuple_146 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_137)); if (unlikely(!__pyx_k_tuple_146)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3604; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_146);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_146));
+  __pyx_k_tuple_148 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_137)); if (unlikely(!__pyx_k_tuple_148)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3606; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_148);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_148));
 
-  /* "mtrand.pyx":3606
+  /* "mtrand.pyx":3608
  *                 raise ValueError("p < 0")
  *             elif fp > 1:
  *                 raise ValueError("p > 1")             # <<<<<<<<<<<<<<
  *             return discdd_array_sc(self.internal_state, rk_negative_binomial,
  *                                    size, fn, fp)
  */
-  __pyx_k_tuple_147 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_147)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3606; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_147);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_147));
+  __pyx_k_tuple_149 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_149)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3608; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_149);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_149));
 
-  /* "mtrand.pyx":3615
+  /* "mtrand.pyx":3617
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(n, 0)):
  *             raise ValueError("n <= 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")
  */
-  __pyx_k_tuple_148 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_144)); if (unlikely(!__pyx_k_tuple_148)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3615; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_148);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_148));
+  __pyx_k_tuple_150 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_146)); if (unlikely(!__pyx_k_tuple_150)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3617; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_150);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_150));
 
-  /* "mtrand.pyx":3617
+  /* "mtrand.pyx":3619
  *             raise ValueError("n <= 0")
  *         if np.any(np.less(p, 0)):
  *             raise ValueError("p < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")
  */
-  __pyx_k_tuple_149 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_137)); if (unlikely(!__pyx_k_tuple_149)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3617; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_149);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_149));
+  __pyx_k_tuple_151 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_137)); if (unlikely(!__pyx_k_tuple_151)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3619; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_151);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_151));
 
-  /* "mtrand.pyx":3619
+  /* "mtrand.pyx":3621
  *             raise ValueError("p < 0")
  *         if np.any(np.greater(p, 1)):
  *             raise ValueError("p > 1")             # <<<<<<<<<<<<<<
  *         return discdd_array(self.internal_state, rk_negative_binomial, size,
  *                             on, op)
  */
-  __pyx_k_tuple_150 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_150)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3619; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_150);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_150));
+  __pyx_k_tuple_152 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_139)); if (unlikely(!__pyx_k_tuple_152)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3621; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_152);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_152));
 
-  /* "mtrand.pyx":3680
+  /* "mtrand.pyx":3682
  *         if not PyErr_Occurred():
  *             if lam < 0:
  *                 raise ValueError("lam < 0")             # <<<<<<<<<<<<<<
  *             if lam > self.poisson_lam_max:
  *                 raise ValueError("lam value too large")
  */
-  __pyx_k_tuple_153 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_152)); if (unlikely(!__pyx_k_tuple_153)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3680; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_153);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_153));
+  __pyx_k_tuple_155 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_154)); if (unlikely(!__pyx_k_tuple_155)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3682; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_155);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_155));
 
-  /* "mtrand.pyx":3682
+  /* "mtrand.pyx":3684
  *                 raise ValueError("lam < 0")
  *             if lam > self.poisson_lam_max:
  *                 raise ValueError("lam value too large")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_poisson, size, flam)
  * 
  */
-  __pyx_k_tuple_155 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_154)); if (unlikely(!__pyx_k_tuple_155)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3682; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_155);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_155));
+  __pyx_k_tuple_157 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_156)); if (unlikely(!__pyx_k_tuple_157)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3684; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_157);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_157));
 
-  /* "mtrand.pyx":3689
+  /* "mtrand.pyx":3691
  *         olam = <ndarray>PyArray_FROM_OTF(lam, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(olam, 0)):
  *             raise ValueError("lam < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(olam, self.poisson_lam_max)):
  *             raise ValueError("lam value too large.")
  */
-  __pyx_k_tuple_156 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_152)); if (unlikely(!__pyx_k_tuple_156)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3689; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_156);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_156));
+  __pyx_k_tuple_158 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_154)); if (unlikely(!__pyx_k_tuple_158)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3691; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_158);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_158));
 
-  /* "mtrand.pyx":3691
+  /* "mtrand.pyx":3693
  *             raise ValueError("lam < 0")
  *         if np.any(np.greater(olam, self.poisson_lam_max)):
  *             raise ValueError("lam value too large.")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_poisson, size, olam)
  * 
  */
-  __pyx_k_tuple_158 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_157)); if (unlikely(!__pyx_k_tuple_158)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3691; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_158);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_158));
+  __pyx_k_tuple_160 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_159)); if (unlikely(!__pyx_k_tuple_160)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3693; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_160);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_160));
 
-  /* "mtrand.pyx":3772
+  /* "mtrand.pyx":3774
  *         if not PyErr_Occurred():
  *             if fa <= 1.0:
  *                 raise ValueError("a <= 1.0")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_zipf, size, fa)
  * 
  */
-  __pyx_k_tuple_160 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_159)); if (unlikely(!__pyx_k_tuple_160)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3772; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_160);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_160));
+  __pyx_k_tuple_162 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_161)); if (unlikely(!__pyx_k_tuple_162)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3774; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_162);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_162));
 
-  /* "mtrand.pyx":3779
+  /* "mtrand.pyx":3781
  *         oa = <ndarray>PyArray_FROM_OTF(a, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(oa, 1.0)):
  *             raise ValueError("a <= 1.0")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_zipf, size, oa)
  * 
  */
-  __pyx_k_tuple_161 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_159)); if (unlikely(!__pyx_k_tuple_161)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3779; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_161);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_161));
+  __pyx_k_tuple_163 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_161)); if (unlikely(!__pyx_k_tuple_163)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3781; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_163);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_163));
 
-  /* "mtrand.pyx":3833
+  /* "mtrand.pyx":3835
  *         if not PyErr_Occurred():
  *             if fp < 0.0:
  *                 raise ValueError("p < 0.0")             # <<<<<<<<<<<<<<
  *             if fp > 1.0:
  *                 raise ValueError("p > 1.0")
  */
-  __pyx_k_tuple_163 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_162)); if (unlikely(!__pyx_k_tuple_163)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3833; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_163);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_163));
+  __pyx_k_tuple_165 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_164)); if (unlikely(!__pyx_k_tuple_165)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3835; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_165);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_165));
 
-  /* "mtrand.pyx":3835
+  /* "mtrand.pyx":3837
  *                 raise ValueError("p < 0.0")
  *             if fp > 1.0:
  *                 raise ValueError("p > 1.0")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_geometric, size, fp)
  * 
  */
-  __pyx_k_tuple_165 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_164)); if (unlikely(!__pyx_k_tuple_165)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3835; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_165);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_165));
+  __pyx_k_tuple_167 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_166)); if (unlikely(!__pyx_k_tuple_167)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3837; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_167);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_167));
 
-  /* "mtrand.pyx":3843
+  /* "mtrand.pyx":3845
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(op, 0.0)):
  *             raise ValueError("p < 0.0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater(op, 1.0)):
  *             raise ValueError("p > 1.0")
  */
-  __pyx_k_tuple_166 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_162)); if (unlikely(!__pyx_k_tuple_166)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3843; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_166);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_166));
+  __pyx_k_tuple_168 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_164)); if (unlikely(!__pyx_k_tuple_168)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3845; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_168);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_168));
 
-  /* "mtrand.pyx":3845
+  /* "mtrand.pyx":3847
  *             raise ValueError("p < 0.0")
  *         if np.any(np.greater(op, 1.0)):
  *             raise ValueError("p > 1.0")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_geometric, size, op)
  * 
  */
-  __pyx_k_tuple_167 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_164)); if (unlikely(!__pyx_k_tuple_167)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3845; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_167);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_167));
+  __pyx_k_tuple_169 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_166)); if (unlikely(!__pyx_k_tuple_169)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3847; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_169);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_169));
 
-  /* "mtrand.pyx":3941
+  /* "mtrand.pyx":3943
  *         if not PyErr_Occurred():
  *             if lngood < 0:
  *                 raise ValueError("ngood < 0")             # <<<<<<<<<<<<<<
  *             if lnbad < 0:
  *                 raise ValueError("nbad < 0")
  */
-  __pyx_k_tuple_169 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_168)); if (unlikely(!__pyx_k_tuple_169)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3941; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_169);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_169));
+  __pyx_k_tuple_171 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_170)); if (unlikely(!__pyx_k_tuple_171)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3943; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_171);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_171));
 
-  /* "mtrand.pyx":3943
+  /* "mtrand.pyx":3945
  *                 raise ValueError("ngood < 0")
  *             if lnbad < 0:
  *                 raise ValueError("nbad < 0")             # <<<<<<<<<<<<<<
  *             if lnsample < 1:
  *                 raise ValueError("nsample < 1")
  */
-  __pyx_k_tuple_171 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_170)); if (unlikely(!__pyx_k_tuple_171)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3943; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_171);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_171));
+  __pyx_k_tuple_173 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_172)); if (unlikely(!__pyx_k_tuple_173)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3945; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_173);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_173));
 
-  /* "mtrand.pyx":3945
+  /* "mtrand.pyx":3947
  *                 raise ValueError("nbad < 0")
  *             if lnsample < 1:
  *                 raise ValueError("nsample < 1")             # <<<<<<<<<<<<<<
  *             if lngood + lnbad < lnsample:
  *                 raise ValueError("ngood + nbad < nsample")
  */
-  __pyx_k_tuple_173 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_172)); if (unlikely(!__pyx_k_tuple_173)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3945; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_173);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_173));
+  __pyx_k_tuple_175 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_174)); if (unlikely(!__pyx_k_tuple_175)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3947; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_175);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_175));
 
-  /* "mtrand.pyx":3947
+  /* "mtrand.pyx":3949
  *                 raise ValueError("nsample < 1")
  *             if lngood + lnbad < lnsample:
  *                 raise ValueError("ngood + nbad < nsample")             # <<<<<<<<<<<<<<
  *             return discnmN_array_sc(self.internal_state, rk_hypergeometric, size,
  *                                     lngood, lnbad, lnsample)
  */
-  __pyx_k_tuple_175 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_174)); if (unlikely(!__pyx_k_tuple_175)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3947; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_175);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_175));
+  __pyx_k_tuple_177 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_176)); if (unlikely(!__pyx_k_tuple_177)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3949; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_177);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_177));
 
-  /* "mtrand.pyx":3957
+  /* "mtrand.pyx":3959
  *         onsample = <ndarray>PyArray_FROM_OTF(nsample, NPY_LONG, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less(ongood, 0)):
  *             raise ValueError("ngood < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(onbad, 0)):
  *             raise ValueError("nbad < 0")
  */
-  __pyx_k_tuple_176 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_168)); if (unlikely(!__pyx_k_tuple_176)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3957; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_176);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_176));
+  __pyx_k_tuple_178 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_170)); if (unlikely(!__pyx_k_tuple_178)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3959; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_178);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_178));
 
-  /* "mtrand.pyx":3959
+  /* "mtrand.pyx":3961
  *             raise ValueError("ngood < 0")
  *         if np.any(np.less(onbad, 0)):
  *             raise ValueError("nbad < 0")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(onsample, 1)):
  *             raise ValueError("nsample < 1")
  */
-  __pyx_k_tuple_177 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_170)); if (unlikely(!__pyx_k_tuple_177)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3959; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_177);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_177));
+  __pyx_k_tuple_179 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_172)); if (unlikely(!__pyx_k_tuple_179)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3961; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_179);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_179));
 
-  /* "mtrand.pyx":3961
+  /* "mtrand.pyx":3963
  *             raise ValueError("nbad < 0")
  *         if np.any(np.less(onsample, 1)):
  *             raise ValueError("nsample < 1")             # <<<<<<<<<<<<<<
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):
  *             raise ValueError("ngood + nbad < nsample")
  */
-  __pyx_k_tuple_178 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_172)); if (unlikely(!__pyx_k_tuple_178)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3961; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_178);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_178));
+  __pyx_k_tuple_180 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_174)); if (unlikely(!__pyx_k_tuple_180)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3963; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_180);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_180));
 
-  /* "mtrand.pyx":3963
+  /* "mtrand.pyx":3965
  *             raise ValueError("nsample < 1")
  *         if np.any(np.less(np.add(ongood, onbad),onsample)):
  *             raise ValueError("ngood + nbad < nsample")             # <<<<<<<<<<<<<<
  *         return discnmN_array(self.internal_state, rk_hypergeometric, size,
  *             ongood, onbad, onsample)
  */
-  __pyx_k_tuple_179 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_174)); if (unlikely(!__pyx_k_tuple_179)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3963; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_179);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_179));
+  __pyx_k_tuple_181 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_176)); if (unlikely(!__pyx_k_tuple_181)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3965; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_181);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_181));
 
-  /* "mtrand.pyx":4047
+  /* "mtrand.pyx":4049
  *         if not PyErr_Occurred():
  *             if fp <= 0.0:
  *                 raise ValueError("p <= 0.0")             # <<<<<<<<<<<<<<
  *             if fp >= 1.0:
  *                 raise ValueError("p >= 1.0")
  */
-  __pyx_k_tuple_181 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_180)); if (unlikely(!__pyx_k_tuple_181)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4047; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_181);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_181));
+  __pyx_k_tuple_183 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_182)); if (unlikely(!__pyx_k_tuple_183)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4049; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_183);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_183));
 
-  /* "mtrand.pyx":4049
+  /* "mtrand.pyx":4051
  *                 raise ValueError("p <= 0.0")
  *             if fp >= 1.0:
  *                 raise ValueError("p >= 1.0")             # <<<<<<<<<<<<<<
  *             return discd_array_sc(self.internal_state, rk_logseries, size, fp)
  * 
  */
-  __pyx_k_tuple_183 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_182)); if (unlikely(!__pyx_k_tuple_183)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4049; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_183);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_183));
+  __pyx_k_tuple_185 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_184)); if (unlikely(!__pyx_k_tuple_185)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4051; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_185);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_185));
 
-  /* "mtrand.pyx":4056
+  /* "mtrand.pyx":4058
  *         op = <ndarray>PyArray_FROM_OTF(p, NPY_DOUBLE, NPY_ARRAY_ALIGNED)
  *         if np.any(np.less_equal(op, 0.0)):
  *             raise ValueError("p <= 0.0")             # <<<<<<<<<<<<<<
  *         if np.any(np.greater_equal(op, 1.0)):
  *             raise ValueError("p >= 1.0")
  */
-  __pyx_k_tuple_184 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_180)); if (unlikely(!__pyx_k_tuple_184)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4056; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_184);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_184));
+  __pyx_k_tuple_186 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_182)); if (unlikely(!__pyx_k_tuple_186)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4058; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_186);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_186));
 
-  /* "mtrand.pyx":4058
+  /* "mtrand.pyx":4060
  *             raise ValueError("p <= 0.0")
  *         if np.any(np.greater_equal(op, 1.0)):
  *             raise ValueError("p >= 1.0")             # <<<<<<<<<<<<<<
  *         return discd_array(self.internal_state, rk_logseries, size, op)
  * 
  */
-  __pyx_k_tuple_185 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_182)); if (unlikely(!__pyx_k_tuple_185)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4058; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_185);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_185));
+  __pyx_k_tuple_187 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_184)); if (unlikely(!__pyx_k_tuple_187)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4060; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_187);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_187));
 
-  /* "mtrand.pyx":4166
+  /* "mtrand.pyx":4168
  * 
  *         if len(mean.shape) != 1:
  *                raise ValueError("mean must be 1 dimensional")             # <<<<<<<<<<<<<<
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):
  *                raise ValueError("cov must be 2 dimensional and square")
  */
-  __pyx_k_tuple_188 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_187)); if (unlikely(!__pyx_k_tuple_188)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4166; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_188);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_188));
+  __pyx_k_tuple_190 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_189)); if (unlikely(!__pyx_k_tuple_190)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4168; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_190);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_190));
 
-  /* "mtrand.pyx":4168
+  /* "mtrand.pyx":4170
  *                raise ValueError("mean must be 1 dimensional")
  *         if (len(cov.shape) != 2) or (cov.shape[0] != cov.shape[1]):
  *                raise ValueError("cov must be 2 dimensional and square")             # <<<<<<<<<<<<<<
  *         if mean.shape[0] != cov.shape[0]:
  *                raise ValueError("mean and cov must have same length")
  */
-  __pyx_k_tuple_190 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_189)); if (unlikely(!__pyx_k_tuple_190)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4168; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_190);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_190));
+  __pyx_k_tuple_192 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_191)); if (unlikely(!__pyx_k_tuple_192)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4170; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_192);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_192));
 
-  /* "mtrand.pyx":4170
+  /* "mtrand.pyx":4172
  *                raise ValueError("cov must be 2 dimensional and square")
  *         if mean.shape[0] != cov.shape[0]:
  *                raise ValueError("mean and cov must have same length")             # <<<<<<<<<<<<<<
  * 
  *         # Compute shape of output and create a matrix of independent
  */
-  __pyx_k_tuple_192 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_191)); if (unlikely(!__pyx_k_tuple_192)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4170; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_192);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_192));
+  __pyx_k_tuple_194 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_193)); if (unlikely(!__pyx_k_tuple_194)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4172; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_194);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_194));
 
-  /* "mtrand.pyx":4176
+  /* "mtrand.pyx":4178
  *         # with the same length as mean and as many rows are necessary to
  *         # form a matrix of shape final_shape.
  *         final_shape = list(shape[:])             # <<<<<<<<<<<<<<
  *         final_shape.append(mean.shape[0])
  *         x = self.standard_normal(final_shape).reshape(-1, mean.shape[0])
  */
-  __pyx_k_slice_193 = PySlice_New(Py_None, Py_None, Py_None); if (unlikely(!__pyx_k_slice_193)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4176; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_slice_193);
-  __Pyx_GIVEREF(__pyx_k_slice_193);
+  __pyx_k_slice_195 = PySlice_New(Py_None, Py_None, Py_None); if (unlikely(!__pyx_k_slice_195)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4178; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_slice_195);
+  __Pyx_GIVEREF(__pyx_k_slice_195);
 
-  /* "mtrand.pyx":4200
+  /* "mtrand.pyx":4202
  *                           RuntimeWarning)
  * 
  *         x = np.dot(x, np.sqrt(s)[:, None] * v)             # <<<<<<<<<<<<<<
  *         x += mean
  *         x.shape = tuple(final_shape)
  */
-  __pyx_k_slice_195 = PySlice_New(Py_None, Py_None, Py_None); if (unlikely(!__pyx_k_slice_195)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_slice_195);
-  __Pyx_GIVEREF(__pyx_k_slice_195);
-  __pyx_k_tuple_196 = PyTuple_Pack(2, __pyx_k_slice_195, Py_None); if (unlikely(!__pyx_k_tuple_196)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4200; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_196);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_196));
+  __pyx_k_slice_197 = PySlice_New(Py_None, Py_None, Py_None); if (unlikely(!__pyx_k_slice_197)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_slice_197);
+  __Pyx_GIVEREF(__pyx_k_slice_197);
+  __pyx_k_tuple_198 = PyTuple_Pack(2, __pyx_k_slice_197, Py_None); if (unlikely(!__pyx_k_tuple_198)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4202; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_198);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_198));
 
-  /* "mtrand.pyx":4269
+  /* "mtrand.pyx":4271
  * 
  *         if kahan_sum(pix, d-1) > (1.0 + 1e-12):
  *             raise ValueError("sum(pvals[:-1]) > 1.0")             # <<<<<<<<<<<<<<
  * 
  *         shape = _shape_from_size(size, d)
  */
-  __pyx_k_tuple_198 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_197)); if (unlikely(!__pyx_k_tuple_198)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4269; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_198);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_198));
+  __pyx_k_tuple_200 = PyTuple_Pack(1, ((PyObject *)__pyx_kp_s_199)); if (unlikely(!__pyx_k_tuple_200)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4271; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_200);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_200));
 
   /* "mtrand.pyx":526
  *     return sum
@@ -23237,10 +23311,10 @@ static int __Pyx_InitCachedConstants(void) {
  *     if size is None:
  *         shape = (d,)
  */
-  __pyx_k_tuple_199 = PyTuple_Pack(3, ((PyObject *)__pyx_n_s__size), ((PyObject *)__pyx_n_s__d), ((PyObject *)__pyx_n_s__shape)); if (unlikely(!__pyx_k_tuple_199)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_199);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_199));
-  __pyx_k_codeobj_200 = (PyObject*)__Pyx_PyCode_New(2, 0, 3, 0, 0, __pyx_empty_bytes, __pyx_empty_tuple, __pyx_empty_tuple, __pyx_k_tuple_199, __pyx_empty_tuple, __pyx_empty_tuple, __pyx_kp_s_201, __pyx_n_s___shape_from_size, 526, __pyx_empty_bytes); if (unlikely(!__pyx_k_codeobj_200)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_k_tuple_201 = PyTuple_Pack(3, ((PyObject *)__pyx_n_s__size), ((PyObject *)__pyx_n_s__d), ((PyObject *)__pyx_n_s__shape)); if (unlikely(!__pyx_k_tuple_201)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_201);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_201));
+  __pyx_k_codeobj_202 = (PyObject*)__Pyx_PyCode_New(2, 0, 3, 0, 0, __pyx_empty_bytes, __pyx_empty_tuple, __pyx_empty_tuple, __pyx_k_tuple_201, __pyx_empty_tuple, __pyx_empty_tuple, __pyx_kp_s_203, __pyx_n_s___shape_from_size, 526, __pyx_empty_bytes); if (unlikely(!__pyx_k_codeobj_202)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
 
   /* "mtrand.pyx":570
  *     """
@@ -23249,12 +23323,12 @@ static int __Pyx_InitCachedConstants(void) {
  * 
  *     def __init__(self, seed=None):
  */
-  __pyx_k_tuple_202 = PyTuple_Pack(1, ((PyObject *)__pyx_n_s__l)); if (unlikely(!__pyx_k_tuple_202)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_202);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_202));
-  __pyx_k_tuple_203 = PyTuple_Pack(1, ((PyObject *)__pyx_n_s__l)); if (unlikely(!__pyx_k_tuple_203)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __Pyx_GOTREF(__pyx_k_tuple_203);
-  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_203));
+  __pyx_k_tuple_204 = PyTuple_Pack(1, ((PyObject *)__pyx_n_s__l)); if (unlikely(!__pyx_k_tuple_204)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_204);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_204));
+  __pyx_k_tuple_205 = PyTuple_Pack(1, ((PyObject *)__pyx_n_s__l)); if (unlikely(!__pyx_k_tuple_205)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __Pyx_GOTREF(__pyx_k_tuple_205);
+  __Pyx_GIVEREF(((PyObject *)__pyx_k_tuple_205));
   __Pyx_RefNannyFinishContext();
   return 0;
   __pyx_L1_error:;
@@ -23359,13 +23433,13 @@ PyMODINIT_FUNC PyInit_mtrand(void)
   /*--- Variable export code ---*/
   /*--- Function export code ---*/
   /*--- Type init code ---*/
+  __pyx_ptype_6mtrand_dtype = __Pyx_ImportType("numpy", "dtype", sizeof(PyArray_Descr), 0); if (unlikely(!__pyx_ptype_6mtrand_dtype)) {__pyx_filename = __pyx_f[1]; __pyx_lineno = 76; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_ptype_6mtrand_ndarray = __Pyx_ImportType("numpy", "ndarray", sizeof(PyArrayObject), 0); if (unlikely(!__pyx_ptype_6mtrand_ndarray)) {__pyx_filename = __pyx_f[1]; __pyx_lineno = 78; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_ptype_6mtrand_flatiter = __Pyx_ImportType("numpy", "flatiter", sizeof(PyArrayIterObject), 0); if (unlikely(!__pyx_ptype_6mtrand_flatiter)) {__pyx_filename = __pyx_f[1]; __pyx_lineno = 80; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_ptype_6mtrand_broadcast = __Pyx_ImportType("numpy", "broadcast", sizeof(PyArrayMultiIterObject), 0); if (unlikely(!__pyx_ptype_6mtrand_broadcast)) {__pyx_filename = __pyx_f[1]; __pyx_lineno = 86; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyType_Ready(&__pyx_type_6mtrand_RandomState) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (__Pyx_SetAttrString(__pyx_m, "RandomState", (PyObject *)&__pyx_type_6mtrand_RandomState) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __pyx_ptype_6mtrand_RandomState = &__pyx_type_6mtrand_RandomState;
-  __pyx_ptype_6mtrand_flatiter = __Pyx_ImportType("numpy", "flatiter", sizeof(PyArrayIterObject), 0); if (unlikely(!__pyx_ptype_6mtrand_flatiter)) {__pyx_filename = __pyx_f[1]; __pyx_lineno = 80; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
-  __pyx_ptype_6mtrand_dtype = __Pyx_ImportType("numpy", "dtype", sizeof(PyArray_Descr), 0); if (unlikely(!__pyx_ptype_6mtrand_dtype)) {__pyx_filename = __pyx_f[1]; __pyx_lineno = 76; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   /*--- Type import code ---*/
   /*--- Variable import code ---*/
   /*--- Function import code ---*/
@@ -23440,7 +23514,7 @@ PyMODINIT_FUNC PyInit_mtrand(void)
   __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__iinfo); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_2);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_1 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_k_tuple_202), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Call(__pyx_t_2, ((PyObject *)__pyx_k_tuple_204), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_2); __pyx_t_2 = 0;
   __pyx_t_2 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__max); if (unlikely(!__pyx_t_2)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
@@ -23456,7 +23530,7 @@ PyMODINIT_FUNC PyInit_mtrand(void)
   __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__iinfo); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  __pyx_t_1 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_k_tuple_203), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = PyObject_Call(__pyx_t_4, ((PyObject *)__pyx_k_tuple_205), NULL); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
   __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__max); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 570; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
@@ -23642,731 +23716,731 @@ PyMODINIT_FUNC PyInit_mtrand(void)
   __Pyx_GIVEREF(__pyx_t_4);
   __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":3623
+  /* "mtrand.pyx":3625
  *                             on, op)
  * 
  *     def poisson(self, lam=1.0, size=None):             # <<<<<<<<<<<<<<
  *         """
  *         poisson(lam=1.0, size=None)
  */
-  __pyx_t_4 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3623; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyFloat_FromDouble(1.0); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 3625; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_k_151 = __pyx_t_4;
+  __pyx_k_153 = __pyx_t_4;
   __Pyx_GIVEREF(__pyx_t_4);
   __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4507
+  /* "mtrand.pyx":4509
  *         return arr
  * 
  * _rand = RandomState()             # <<<<<<<<<<<<<<
  * seed = _rand.seed
  * get_state = _rand.get_state
  */
-  __pyx_t_4 = PyObject_Call(((PyObject *)((PyObject*)__pyx_ptype_6mtrand_RandomState)), ((PyObject *)__pyx_empty_tuple), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4507; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = PyObject_Call(((PyObject *)((PyObject*)__pyx_ptype_6mtrand_RandomState)), ((PyObject *)__pyx_empty_tuple), NULL); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4509; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s___rand, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4507; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s___rand, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4509; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4508
+  /* "mtrand.pyx":4510
  * 
  * _rand = RandomState()
  * seed = _rand.seed             # <<<<<<<<<<<<<<
  * get_state = _rand.get_state
  * set_state = _rand.set_state
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4508; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4510; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__seed); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4508; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__seed); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4510; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__seed, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4508; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__seed, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4510; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4509
+  /* "mtrand.pyx":4511
  * _rand = RandomState()
  * seed = _rand.seed
  * get_state = _rand.get_state             # <<<<<<<<<<<<<<
  * set_state = _rand.set_state
  * random_sample = _rand.random_sample
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4509; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4511; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__get_state); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4509; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__get_state); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4511; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__get_state, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4509; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__get_state, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4511; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4510
+  /* "mtrand.pyx":4512
  * seed = _rand.seed
  * get_state = _rand.get_state
  * set_state = _rand.set_state             # <<<<<<<<<<<<<<
  * random_sample = _rand.random_sample
  * choice = _rand.choice
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4510; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4512; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__set_state); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4510; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__set_state); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4512; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__set_state, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4510; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__set_state, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4512; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4511
+  /* "mtrand.pyx":4513
  * get_state = _rand.get_state
  * set_state = _rand.set_state
  * random_sample = _rand.random_sample             # <<<<<<<<<<<<<<
  * choice = _rand.choice
  * randint = _rand.randint
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4511; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__random_sample); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4511; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__random_sample); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__random_sample, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4511; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__random_sample, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4512
+  /* "mtrand.pyx":4514
  * set_state = _rand.set_state
  * random_sample = _rand.random_sample
  * choice = _rand.choice             # <<<<<<<<<<<<<<
  * randint = _rand.randint
  * bytes = _rand.bytes
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4512; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__choice); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4512; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__choice); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__choice, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4512; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__choice, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4513
+  /* "mtrand.pyx":4515
  * random_sample = _rand.random_sample
  * choice = _rand.choice
  * randint = _rand.randint             # <<<<<<<<<<<<<<
  * bytes = _rand.bytes
  * uniform = _rand.uniform
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__randint); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__randint); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__randint, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4513; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__randint, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4514
+  /* "mtrand.pyx":4516
  * choice = _rand.choice
  * randint = _rand.randint
  * bytes = _rand.bytes             # <<<<<<<<<<<<<<
  * uniform = _rand.uniform
  * rand = _rand.rand
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__bytes); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__bytes); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__bytes, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4514; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__bytes, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4515
+  /* "mtrand.pyx":4517
  * randint = _rand.randint
  * bytes = _rand.bytes
  * uniform = _rand.uniform             # <<<<<<<<<<<<<<
  * rand = _rand.rand
  * randn = _rand.randn
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4517; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__uniform); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__uniform); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4517; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__uniform, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4515; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__uniform, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4517; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4516
+  /* "mtrand.pyx":4518
  * bytes = _rand.bytes
  * uniform = _rand.uniform
  * rand = _rand.rand             # <<<<<<<<<<<<<<
  * randn = _rand.randn
  * random_integers = _rand.random_integers
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__rand, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4516; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__rand, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4517
+  /* "mtrand.pyx":4519
  * uniform = _rand.uniform
  * rand = _rand.rand
  * randn = _rand.randn             # <<<<<<<<<<<<<<
  * random_integers = _rand.random_integers
  * standard_normal = _rand.standard_normal
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4517; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__randn); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4517; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__randn); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__randn, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4517; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__randn, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4518
+  /* "mtrand.pyx":4520
  * rand = _rand.rand
  * randn = _rand.randn
  * random_integers = _rand.random_integers             # <<<<<<<<<<<<<<
  * standard_normal = _rand.standard_normal
  * normal = _rand.normal
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__random_integers); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__random_integers); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__random_integers, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4518; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__random_integers, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4519
+  /* "mtrand.pyx":4521
  * randn = _rand.randn
  * random_integers = _rand.random_integers
  * standard_normal = _rand.standard_normal             # <<<<<<<<<<<<<<
  * normal = _rand.normal
  * beta = _rand.beta
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__standard_normal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__standard_normal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_normal, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4519; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_normal, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4520
+  /* "mtrand.pyx":4522
  * random_integers = _rand.random_integers
  * standard_normal = _rand.standard_normal
  * normal = _rand.normal             # <<<<<<<<<<<<<<
  * beta = _rand.beta
  * exponential = _rand.exponential
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__normal); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__normal); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__normal, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4520; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__normal, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4521
+  /* "mtrand.pyx":4523
  * standard_normal = _rand.standard_normal
  * normal = _rand.normal
  * beta = _rand.beta             # <<<<<<<<<<<<<<
  * exponential = _rand.exponential
  * standard_exponential = _rand.standard_exponential
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__beta); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__beta); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__beta, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4521; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__beta, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4522
+  /* "mtrand.pyx":4524
  * normal = _rand.normal
  * beta = _rand.beta
  * exponential = _rand.exponential             # <<<<<<<<<<<<<<
  * standard_exponential = _rand.standard_exponential
  * standard_gamma = _rand.standard_gamma
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__exponential); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__exponential); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__exponential, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4522; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__exponential, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4523
+  /* "mtrand.pyx":4525
  * beta = _rand.beta
  * exponential = _rand.exponential
  * standard_exponential = _rand.standard_exponential             # <<<<<<<<<<<<<<
  * standard_gamma = _rand.standard_gamma
  * gamma = _rand.gamma
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s_204); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s_206); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s_204, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4523; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s_206, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4524
+  /* "mtrand.pyx":4526
  * exponential = _rand.exponential
  * standard_exponential = _rand.standard_exponential
  * standard_gamma = _rand.standard_gamma             # <<<<<<<<<<<<<<
  * gamma = _rand.gamma
  * f = _rand.f
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__standard_gamma); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__standard_gamma); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_gamma, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4524; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_gamma, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4525
+  /* "mtrand.pyx":4527
  * standard_exponential = _rand.standard_exponential
  * standard_gamma = _rand.standard_gamma
  * gamma = _rand.gamma             # <<<<<<<<<<<<<<
  * f = _rand.f
  * noncentral_f = _rand.noncentral_f
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__gamma); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__gamma); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__gamma, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4525; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__gamma, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4526
+  /* "mtrand.pyx":4528
  * standard_gamma = _rand.standard_gamma
  * gamma = _rand.gamma
  * f = _rand.f             # <<<<<<<<<<<<<<
  * noncentral_f = _rand.noncentral_f
  * chisquare = _rand.chisquare
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__f); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__f); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__f, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4526; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__f, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4527
+  /* "mtrand.pyx":4529
  * gamma = _rand.gamma
  * f = _rand.f
  * noncentral_f = _rand.noncentral_f             # <<<<<<<<<<<<<<
  * chisquare = _rand.chisquare
  * noncentral_chisquare = _rand.noncentral_chisquare
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4529; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__noncentral_f); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__noncentral_f); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4529; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__noncentral_f, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4527; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__noncentral_f, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4529; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4528
+  /* "mtrand.pyx":4530
  * f = _rand.f
  * noncentral_f = _rand.noncentral_f
  * chisquare = _rand.chisquare             # <<<<<<<<<<<<<<
  * noncentral_chisquare = _rand.noncentral_chisquare
  * standard_cauchy = _rand.standard_cauchy
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4530; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__chisquare); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__chisquare); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4530; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__chisquare, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4528; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__chisquare, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4530; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4529
+  /* "mtrand.pyx":4531
  * noncentral_f = _rand.noncentral_f
  * chisquare = _rand.chisquare
  * noncentral_chisquare = _rand.noncentral_chisquare             # <<<<<<<<<<<<<<
  * standard_cauchy = _rand.standard_cauchy
  * standard_t = _rand.standard_t
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4529; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4531; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s_205); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4529; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s_207); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4531; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s_205, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4529; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s_207, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4531; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4530
+  /* "mtrand.pyx":4532
  * chisquare = _rand.chisquare
  * noncentral_chisquare = _rand.noncentral_chisquare
  * standard_cauchy = _rand.standard_cauchy             # <<<<<<<<<<<<<<
  * standard_t = _rand.standard_t
  * vonmises = _rand.vonmises
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4530; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__standard_cauchy); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4530; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__standard_cauchy); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_cauchy, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4530; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_cauchy, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4531
+  /* "mtrand.pyx":4533
  * noncentral_chisquare = _rand.noncentral_chisquare
  * standard_cauchy = _rand.standard_cauchy
  * standard_t = _rand.standard_t             # <<<<<<<<<<<<<<
  * vonmises = _rand.vonmises
  * pareto = _rand.pareto
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4531; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4533; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__standard_t); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4531; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__standard_t); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4533; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_t, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4531; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__standard_t, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4533; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4532
+  /* "mtrand.pyx":4534
  * standard_cauchy = _rand.standard_cauchy
  * standard_t = _rand.standard_t
  * vonmises = _rand.vonmises             # <<<<<<<<<<<<<<
  * pareto = _rand.pareto
  * weibull = _rand.weibull
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4534; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__vonmises); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__vonmises); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4534; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__vonmises, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4532; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__vonmises, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4534; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4533
+  /* "mtrand.pyx":4535
  * standard_t = _rand.standard_t
  * vonmises = _rand.vonmises
  * pareto = _rand.pareto             # <<<<<<<<<<<<<<
  * weibull = _rand.weibull
  * power = _rand.power
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4533; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4535; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__pareto); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4533; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__pareto); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4535; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__pareto, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4533; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__pareto, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4535; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4534
+  /* "mtrand.pyx":4536
  * vonmises = _rand.vonmises
  * pareto = _rand.pareto
  * weibull = _rand.weibull             # <<<<<<<<<<<<<<
  * power = _rand.power
  * laplace = _rand.laplace
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4534; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__weibull); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4534; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__weibull); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__weibull, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4534; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__weibull, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4535
+  /* "mtrand.pyx":4537
  * pareto = _rand.pareto
  * weibull = _rand.weibull
  * power = _rand.power             # <<<<<<<<<<<<<<
  * laplace = _rand.laplace
  * gumbel = _rand.gumbel
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4535; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4537; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__power); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4535; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__power); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4537; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__power, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4535; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__power, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4537; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4536
+  /* "mtrand.pyx":4538
  * weibull = _rand.weibull
  * power = _rand.power
  * laplace = _rand.laplace             # <<<<<<<<<<<<<<
  * gumbel = _rand.gumbel
  * logistic = _rand.logistic
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4538; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__laplace); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__laplace); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4538; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__laplace, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4536; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__laplace, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4538; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4537
+  /* "mtrand.pyx":4539
  * power = _rand.power
  * laplace = _rand.laplace
  * gumbel = _rand.gumbel             # <<<<<<<<<<<<<<
  * logistic = _rand.logistic
  * lognormal = _rand.lognormal
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4537; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4539; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__gumbel); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4537; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__gumbel); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4539; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__gumbel, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4537; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__gumbel, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4539; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4538
+  /* "mtrand.pyx":4540
  * laplace = _rand.laplace
  * gumbel = _rand.gumbel
  * logistic = _rand.logistic             # <<<<<<<<<<<<<<
  * lognormal = _rand.lognormal
  * rayleigh = _rand.rayleigh
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4538; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4540; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__logistic); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4538; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__logistic); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4540; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__logistic, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4538; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__logistic, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4540; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4539
+  /* "mtrand.pyx":4541
  * gumbel = _rand.gumbel
  * logistic = _rand.logistic
  * lognormal = _rand.lognormal             # <<<<<<<<<<<<<<
  * rayleigh = _rand.rayleigh
  * wald = _rand.wald
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4539; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4541; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__lognormal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4539; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__lognormal); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4541; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__lognormal, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4539; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__lognormal, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4541; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4540
+  /* "mtrand.pyx":4542
  * logistic = _rand.logistic
  * lognormal = _rand.lognormal
  * rayleigh = _rand.rayleigh             # <<<<<<<<<<<<<<
  * wald = _rand.wald
  * triangular = _rand.triangular
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4540; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4542; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__rayleigh); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4540; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__rayleigh); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4542; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__rayleigh, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4540; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__rayleigh, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4542; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4541
+  /* "mtrand.pyx":4543
  * lognormal = _rand.lognormal
  * rayleigh = _rand.rayleigh
  * wald = _rand.wald             # <<<<<<<<<<<<<<
  * triangular = _rand.triangular
  * 
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4541; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4543; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__wald); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4541; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__wald); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4543; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__wald, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4541; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__wald, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4543; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4542
+  /* "mtrand.pyx":4544
  * rayleigh = _rand.rayleigh
  * wald = _rand.wald
  * triangular = _rand.triangular             # <<<<<<<<<<<<<<
  * 
  * binomial = _rand.binomial
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4542; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4544; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__triangular); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4542; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__triangular); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4544; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__triangular, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4542; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__triangular, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4544; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4544
+  /* "mtrand.pyx":4546
  * triangular = _rand.triangular
  * 
  * binomial = _rand.binomial             # <<<<<<<<<<<<<<
  * negative_binomial = _rand.negative_binomial
  * poisson = _rand.poisson
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4544; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4546; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__binomial); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4544; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__binomial); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4546; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__binomial, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4544; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__binomial, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4546; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4545
+  /* "mtrand.pyx":4547
  * 
  * binomial = _rand.binomial
  * negative_binomial = _rand.negative_binomial             # <<<<<<<<<<<<<<
  * poisson = _rand.poisson
  * zipf = _rand.zipf
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4545; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4547; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__negative_binomial); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4545; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__negative_binomial); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4547; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__negative_binomial, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4545; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__negative_binomial, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4547; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4546
+  /* "mtrand.pyx":4548
  * binomial = _rand.binomial
  * negative_binomial = _rand.negative_binomial
  * poisson = _rand.poisson             # <<<<<<<<<<<<<<
  * zipf = _rand.zipf
  * geometric = _rand.geometric
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4546; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4548; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__poisson); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4546; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__poisson); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4548; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__poisson, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4546; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__poisson, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4548; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4547
+  /* "mtrand.pyx":4549
  * negative_binomial = _rand.negative_binomial
  * poisson = _rand.poisson
  * zipf = _rand.zipf             # <<<<<<<<<<<<<<
  * geometric = _rand.geometric
  * hypergeometric = _rand.hypergeometric
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4547; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4549; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__zipf); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4547; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__zipf); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4549; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__zipf, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4547; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__zipf, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4549; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4548
+  /* "mtrand.pyx":4550
  * poisson = _rand.poisson
  * zipf = _rand.zipf
  * geometric = _rand.geometric             # <<<<<<<<<<<<<<
  * hypergeometric = _rand.hypergeometric
  * logseries = _rand.logseries
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4548; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4550; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__geometric); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4548; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__geometric); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4550; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__geometric, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4548; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__geometric, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4550; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4549
+  /* "mtrand.pyx":4551
  * zipf = _rand.zipf
  * geometric = _rand.geometric
  * hypergeometric = _rand.hypergeometric             # <<<<<<<<<<<<<<
  * logseries = _rand.logseries
  * 
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4549; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4551; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__hypergeometric); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4549; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__hypergeometric); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4551; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__hypergeometric, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4549; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__hypergeometric, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4551; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4550
+  /* "mtrand.pyx":4552
  * geometric = _rand.geometric
  * hypergeometric = _rand.hypergeometric
  * logseries = _rand.logseries             # <<<<<<<<<<<<<<
  * 
  * multivariate_normal = _rand.multivariate_normal
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4550; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4552; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__logseries); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4550; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__logseries); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4552; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__logseries, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4550; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__logseries, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4552; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4552
+  /* "mtrand.pyx":4554
  * logseries = _rand.logseries
  * 
  * multivariate_normal = _rand.multivariate_normal             # <<<<<<<<<<<<<<
  * multinomial = _rand.multinomial
  * dirichlet = _rand.dirichlet
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4552; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4554; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__multivariate_normal); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4552; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__multivariate_normal); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4554; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__multivariate_normal, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4552; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__multivariate_normal, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4554; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4553
+  /* "mtrand.pyx":4555
  * 
  * multivariate_normal = _rand.multivariate_normal
  * multinomial = _rand.multinomial             # <<<<<<<<<<<<<<
  * dirichlet = _rand.dirichlet
  * 
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4553; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4555; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__multinomial); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4553; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__multinomial); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4555; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__multinomial, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4553; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__multinomial, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4555; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4554
+  /* "mtrand.pyx":4556
  * multivariate_normal = _rand.multivariate_normal
  * multinomial = _rand.multinomial
  * dirichlet = _rand.dirichlet             # <<<<<<<<<<<<<<
  * 
  * shuffle = _rand.shuffle
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4554; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4556; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__dirichlet); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4554; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__dirichlet); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4556; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__dirichlet, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4554; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__dirichlet, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4556; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
-  /* "mtrand.pyx":4556
+  /* "mtrand.pyx":4558
  * dirichlet = _rand.dirichlet
  * 
  * shuffle = _rand.shuffle             # <<<<<<<<<<<<<<
  * permutation = _rand.permutation
  */
-  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4556; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4558; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
-  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__shuffle); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4556; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_PyObject_GetAttrStr(__pyx_t_1, __pyx_n_s__shuffle); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4558; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__shuffle, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4556; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__shuffle, __pyx_t_4) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4558; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
 
-  /* "mtrand.pyx":4557
+  /* "mtrand.pyx":4559
  * 
  * shuffle = _rand.shuffle
  * permutation = _rand.permutation             # <<<<<<<<<<<<<<
  */
-  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4557; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_4 = __Pyx_GetModuleGlobalName(__pyx_n_s___rand); if (unlikely(!__pyx_t_4)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4559; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_4);
-  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__permutation); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4557; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  __pyx_t_1 = __Pyx_PyObject_GetAttrStr(__pyx_t_4, __pyx_n_s__permutation); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4559; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(__pyx_t_1);
   __Pyx_DECREF(__pyx_t_4); __pyx_t_4 = 0;
-  if (PyDict_SetItem(__pyx_d, __pyx_n_s__permutation, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4557; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_d, __pyx_n_s__permutation, __pyx_t_1) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 4559; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(__pyx_t_1); __pyx_t_1 = 0;
 
   /* "mtrand.pyx":1
@@ -24376,7 +24450,6 @@ PyMODINIT_FUNC PyInit_mtrand(void)
  */
   __pyx_t_1 = PyDict_New(); if (unlikely(!__pyx_t_1)) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_GOTREF(((PyObject *)__pyx_t_1));
-  if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_206), ((PyObject *)__pyx_kp_u_207)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_208), ((PyObject *)__pyx_kp_u_209)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_210), ((PyObject *)__pyx_kp_u_211)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_212), ((PyObject *)__pyx_kp_u_213)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
@@ -24419,6 +24492,7 @@ PyMODINIT_FUNC PyInit_mtrand(void)
   if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_286), ((PyObject *)__pyx_kp_u_287)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_288), ((PyObject *)__pyx_kp_u_289)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_290), ((PyObject *)__pyx_kp_u_291)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
+  if (PyDict_SetItem(__pyx_t_1, ((PyObject *)__pyx_kp_u_292), ((PyObject *)__pyx_kp_u_293)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   if (PyDict_SetItem(__pyx_d, __pyx_n_s____test__, ((PyObject *)__pyx_t_1)) < 0) {__pyx_filename = __pyx_f[0]; __pyx_lineno = 1; __pyx_clineno = __LINE__; goto __pyx_L1_error;}
   __Pyx_DECREF(((PyObject *)__pyx_t_1)); __pyx_t_1 = 0;
   goto __pyx_L0;
diff --git a/numpy/random/mtrand/mtrand.pyx b/numpy/random/mtrand/mtrand.pyx
index b33e132a4..7dd73fab7 100644
--- a/numpy/random/mtrand/mtrand.pyx
+++ b/numpy/random/mtrand/mtrand.pyx
@@ -3511,6 +3511,8 @@ cdef class RandomState:
                 raise ValueError("p < 0")
             elif fp > 1:
                 raise ValueError("p > 1")
+            elif np.isnan(fp):
+                raise ValueError("p is nan")
             return discnp_array_sc(self.internal_state, rk_binomial, size, ln, fp)
 
         PyErr_Clear()
diff --git a/numpy/random/tests/test_random.py b/numpy/random/tests/test_random.py
index c30037e1e..db4093ab4 100644
--- a/numpy/random/tests/test_random.py
+++ b/numpy/random/tests/test_random.py
@@ -18,6 +18,10 @@ class TestBinomial(TestCase):
             assert_(random.binomial(0, p) == 0)
             np.testing.assert_array_equal(random.binomial(zeros, p), zeros)
 
+    def test_p_is_nan(self):
+        # Issue #4571.
+        assert_raises(ValueError, random.binomial, 1, np.nan)
+
 
 class TestMultinomial(TestCase):
     def test_basic(self):