Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge pull request #9577 from bashtage/dirichlet-alpha-validation | Charles Harris | 2017-08-17 | 2 | -0/+12 |
|\ | | | | | BUG: Missing dirichlet input validation | ||||
| * | BUG: Missing dirichlet input validation | Kevin Sheppard | 2017-08-17 | 2 | -0/+12 |
| | | | | | | | | | | | | | | | | Dirichlet does not validate inputs and hangs when values are zero. Adds check that values are strictly positive as required by the distribution. closes #2089 | ||||
* | | BUG: core: remove extra return statement | Pauli Virtanen | 2017-08-17 | 1 | -1/+1 |
| | | |||||
* | | BUG: core: blacklist MSVC cabs* on win32 | Pauli Virtanen | 2017-08-17 | 1 | -0/+4 |
| | | | | | | | | | | Similarly to hypot() also cabs* on MSVC/win32 mess with floating point mode. | ||||
* | | TST: core: add a test that exercise untested npymath functions | Pauli Virtanen | 2017-08-17 | 3 | -1/+160 |
|/ | | | | | These functions do not occur currently in numpy codebase either at all or only very indirectly, so they are not tested otherwise. | ||||
* | TST: add FPU mode check also for pytest | Pauli Virtanen | 2017-08-16 | 1 | -0/+54 |
| | |||||
* | ENH: check for FPU mode changes in the test suite | Pauli Virtanen | 2017-08-16 | 3 | -3/+72 |
| | | | | | Emit a test failure if the FPU mode changes when running a test case, allowing to pinpoint what test caused the mode change. | ||||
* | BUG: blacklist MSVC hypot() on win32 | Pauli Virtanen | 2017-08-16 | 1 | -0/+9 |
| | |||||
* | BUG: fix distutils/cpuinfo.py:getoutput() | Mike Nolta | 2017-08-15 | 1 | -1/+1 |
| | | | | | | | | | If getstatusoutput() throws an exception, getoutput() tries to catch it, but then crashes with: UnboundLocalError: local variable 'output' referenced before assignment because it tries to return the non-existent result of getstatusoutput(). | ||||
* | BUG: fix regression in 1.13.x in distutils.mingw32ccompiler. | Ralf Gommers | 2017-08-15 | 1 | -1/+1 |
| | | | | | | Issue was introduced in gh-8454. Thanks to @jennirinker for pointing out the issue and fix. Closes gh-9553. | ||||
* | DOC, BUG: Fix Python 3.6 invalid escape sequence. | Charles Harris | 2017-08-13 | 1 | -2/+2 |
| | | | | | | | | | The rst markup in numpy/doc/basics.py uses `\s`, which is interpreted by python 3.6 as a deprecated escape sequence. Fix by escaping the `\`. Closes #9551. [ci skip] | ||||
* | DOC: add example for isscalar on strings (#9361) | Christoph Boeddeker | 2017-08-13 | 1 | -0/+2 |
| | |||||
* | BUG: MaskedArray._optinfo is discarded by__eq__/__ne__ (#9540) | David Schaefer | 2017-08-13 | 2 | -0/+20 |
| | |||||
* | Merge pull request #9527 from Licht-T/fix-scalar-astype | Eric Wieser | 2017-08-13 | 1 | -13/+14 |
|\ | | | | | BUG: Fix scalar methods to receive keyword arguments | ||||
| * | BUG: Fix scalar-ravel so that it can receive keyword arguments | Licht Takeuchi | 2017-08-10 | 1 | -3/+3 |
| | | |||||
| * | BUG: Fix scalar-flatten so that it can receive keyword arguments | Licht Takeuchi | 2017-08-10 | 1 | -3/+4 |
| | | |||||
| * | BUG: Fix scalar-view so that it can receive keyword arguments | Licht Takeuchi | 2017-08-10 | 1 | -3/+3 |
| | | |||||
| * | BUG: Fix scalar-searchsorted so that it can receive keyword arguments | Licht Takeuchi | 2017-08-10 | 1 | -3/+3 |
| | | |||||
| * | BUG: Fix scalar-copy so that it can receive keyword arguments | Licht Takeuchi | 2017-08-10 | 1 | -3/+3 |
| | | |||||
| * | BUG: Fix scalar-tobytes so that it can receive keyword arguments | Licht Takeuchi | 2017-08-09 | 1 | -3/+3 |
| | | |||||
| * | BUG: Fix scalar-tostring so that it can receive keyword arguments | Licht Takeuchi | 2017-08-09 | 1 | -3/+3 |
| | | |||||
| * | BUG: Fix scalar-astype so that it can receive keyword arguments | Licht Takeuchi | 2017-08-07 | 1 | -4/+4 |
| | | |||||
* | | Merge pull request #9524 from charris/add_init_to_test_directories | Charles Harris | 2017-08-12 | 31 | -18/+18 |
|\ \ | | | | | | | TST, MAINT: Add `__init__.py` files to tests directories. | ||||
| * | | TST, MAINT: Add `__init__.py` files to tests directories. | Charles Harris | 2017-08-06 | 31 | -18/+18 |
| |/ | | | | | | | | | | | | | | | This allows pytest to run with duplicate test file names. Note that `python <path-to-test-file>` no longer works with this change, nor will a simple `pytest numpy`, because numpy is imported from the numpy repository. However, `python runtests.py` and `>>> numpy.test()` are still available. | ||||
* | | Merge branch 'master' into fix-note-on-quad-prec | Charles Harris | 2017-08-12 | 18 | -53/+64 |
|\ \ | |||||
| * \ | Merge pull request #9422 from deniederhut/doc/types-formatting | Charles Harris | 2017-08-12 | 1 | -2/+2 |
| |\ \ | | | | | | | | | DOC: correct formatting of basic.types.html | ||||
| | * | | DOC: reverts some escapes [ci-skip] | Dillon Niederhut | 2017-08-11 | 1 | -2/+2 |
| | | | | |||||
| | * | | DOC: removes extraneous whitespace with escapes | Dillon Niederhut | 2017-07-27 | 1 | -3/+3 |
| | | | | | | | | | | | | | | | | [ci-skip] | ||||
| | * | | DOC: correct formatting of basic.types.html | Dillon Niederhut | 2017-07-26 | 1 | -3/+3 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the documentation for types allowed in numpy, missing spaces around the backticks for fixed-width formatting cause code examples to appear as plain text, or are causing plain text to appear as code. This commit fixes back tick spacing in the 'Extended Precision' section of the 'Data Types' page. | ||||
| * | | | MAINT: Use the error_converting macro where possible | Eric Wieser | 2017-08-11 | 17 | -51/+62 |
| | | | | |||||
* | | | | DOC: Adding backslash between double-backtick and s. | Danny Hermes | 2017-08-11 | 1 | -1/+1 |
|/ / / | |||||
* | | | MAINT: Use moveaxis instead of rollaxis internally (#9475) | Nico Schlömer | 2017-08-10 | 13 | -52/+53 |
| | | | | | | | | | | | | | | | Also add a hint to the documentation advising the use of moveaxis over rollaxis. Tests for rollaxis are left alone. | ||||
* | | | BUG/TST: Fix the polynomial.test_printing tests, which never tested anything ↵ | Eric Wieser | 2017-08-10 | 1 | -25/+25 |
| | | | | | | | | | | | | | | | | | | useful assert_ is not a substitute for assert_equal | ||||
* | | | ENH: Show domain and window as kwargs in repr | Eric Wieser | 2017-08-10 | 3 | -6/+6 |
| | | | | | | | | | | | | Also, update the docs with this new repr | ||||
* | | | BUG: The NAT deprecation warning should not be given for every single value | Sebastian Berg | 2017-08-07 | 2 | -20/+38 |
| |/ |/| | | | | | | | | | | | | | | | This fixes it so that it is at least only given for each inner loop run, which admittingly slightly modifies the output on error (it will not stop at the first NAT, but rather only after the first inner loop finished; this does not relaly bother me though). Closes gh-9528 | ||||
* | | Merge pull request #9522 from eric-wieser/stop-using-obj2sctype | Charles Harris | 2017-08-06 | 4 | -46/+16 |
|\ \ | | | | | | | BUG: Fix problems with obj2sctype | ||||
| * | | MAINT: Remove unneeded asfarray | Eric Wieser | 2017-08-06 | 1 | -12/+1 |
| | | | | | | | | | | | | As of gh-7088, x is always a float array anyway | ||||
| * | | BUG: Don't allow an array to be passed as the dtype argument | Eric Wieser | 2017-08-06 | 2 | -3/+8 |
| | | | |||||
| * | | BUG: Don't duplicate the type->dtype conversion logic | Eric Wieser | 2017-08-06 | 1 | -31/+7 |
| | | | | | | | | | | | | | | | | | | | | | | | | dtype(t) already does this in descriptor.c This means that `obj2sctype(long)` now returns `np.int64` rather than `None`, because when you duplicate code, its easy to forget something in the second place. | ||||
* | | | Merge pull request #9505 from eric-wieser/fix-issubdtype | Charles Harris | 2017-08-06 | 11 | -22/+86 |
|\ \ \ | | | | | | | | | BUG: issubdtype is inconsistent on types and dtypes | ||||
| * | | | DEP: Deprecate the coercion of dtype-likes to their superclass | Eric Wieser | 2017-08-05 | 1 | -0/+22 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In many cases, this coercion is surprising, or would be if the user knew about it: * [('a', int)] -> np.flexible * str - > str (!) - not even a numpy type * 'float32' -> np.floating (discards size) * int -> np.signed_integer (not np.integer, as is usually meant) | ||||
| * | | | DOC: Make example work in python 3 | Eric Wieser | 2017-08-05 | 1 | -1/+1 |
| | | | | |||||
| * | | | BUG: dtypes are ignored and downcast in issubdtype | Eric Wieser | 2017-08-05 | 2 | -8/+36 |
| | | | | | | | | | | | | | | | | Fixes #9480 | ||||
| * | | | BUG: abstract types did not compare correctly with issubdtype | Eric Wieser | 2017-08-05 | 2 | -2/+11 |
| | | | | |||||
| * | | | MAINT: Simplify test for string-likes | Eric Wieser | 2017-08-05 | 1 | -2/+1 |
| | | | | |||||
| * | | | MAINT: Stop using the undocumented coercion-then-downcast feature of subdtype | Eric Wieser | 2017-08-05 | 6 | -10/+10 |
| | | | | |||||
| * | | | BUG: Be explicit about allowing all integer types in matrix exponentiation | Eric Wieser | 2017-08-05 | 2 | -1/+7 |
| | | | | | | | | | | | | | | | | Fixes gh-9506, unsigned exponentiation | ||||
* | | | | Merge pull request #9521 from eric-wieser/numbers.Number | Charles Harris | 2017-08-06 | 2 | -0/+14 |
|\ \ \ \ | |_|/ / |/| | | | BUG: Make issubclass(np.number, numbers.Number) return true | ||||
| * | | | BUG: Make issubclass(np.number, numbers.Number) return true | Eric Wieser | 2017-08-05 | 2 | -0/+14 |
| |/ / | | | | | | | | | | An omission from 2d73ff34f48980778b4e808afc76c0c1fa841ba7 | ||||
* | | | Merge pull request #9520 from eric-wieser/remove-TypeType | seberg | 2017-08-06 | 1 | -12/+6 |
|\ \ \ | | | | | | | | | MAINT: types.TypeType does not ever need to be used |