summaryrefslogtreecommitdiff
path: root/numpy/lib
Commit message (Collapse)AuthorAgeFilesLines
...
| * BUG: Preserve field order in join_by, avoids FutureWarningAllan Haldane2017-05-092-6/+49
| | | | | | | | Fixes #8940
* | ENH: Spelling fixesVille Skyttä2017-05-098-9/+9
|/
* DOC: fix docstring for np.isinStephan Hoyer2017-05-071-0/+1
|
* ENH: add divmod support to NDArrayOperatorsMixinStephan Hoyer2017-05-072-40/+60
|
* BUG: np.ma.mr_['r',...] does not return masked arraysEric Wieser2017-05-051-3/+3
|
* BUG: np.r_['r',...] crashes on scalarsEric Wieser2017-05-052-1/+6
|
* BUG: np.r_['rc'] is treated as np.r_['r'], not an errorEric Wieser2017-05-052-1/+3
|
* TST: Add some tests for passing r and c to r_Eric Wieser2017-05-051-0/+13
|
* MAINT: Use enumerate instead of range(len(...))Eric Wieser2017-05-051-15/+14
|
* BUG: Remove mutable state from AxisConcatenatorEric Wieser2017-05-052-29/+37
| | | | Fixes #8815
* MAINT: Remove code duplicated from np.r_ in np.ma.mr_Eric Wieser2017-05-051-1/+3
| | | | | Also adds a test for the disabled-by-design behaviour - this would return raw matrices, not masked arrays
* ENH: Add isin, genereralizing in1d to ND arrays (#8423)B R S Recht2017-05-053-7/+145
| | | | | | This fixes gh-8331 Also update the docs for arraysetops to remove the outdated "1D" from the description, which was already incorrect for np.unique.
* ENH: add __pos__ to NDArrayOperatorsMixinStephan Hoyer2017-04-302-3/+4
|
* Merge pull request #8885 from juliantaylor/tracemallocCharles Harris2017-04-301-1/+2
|\ | | | | ENH: add support for python3.6 memory tracing
| * ENH: add support for python3.6 memory tracingJulian Taylor2017-04-281-1/+2
| | | | | | | | | | | | | | | | | | Python 3.6 added a private API for tracing arbitrary memory allocations so the tracemalloc module [0] can be used with numpy. closes gh-4663 [0] https://docs.python.org/3/library/tracemalloc.html
* | Merge pull request #9013 from eric-wieser/fix-nanfuncsCharles Harris2017-04-282-15/+31
|\ \ | |/ |/| BUG: Fix np.lib.nanfunctions on object arrays
| * BUG: Fix incorrect behavior of nanfunctions on object arraysEric Wieser2017-04-292-8/+28
| | | | | | | | Fixes gh-8974 and gh-9008
| * MAINT: Remove weird create-a-copy danceEric Wieser2017-04-271-6/+2
| |
* | Merge pull request #8996 from eric-wieser/fix-ufunclikeMarten van Kerkwijk2017-04-272-27/+84
|\ \ | | | | | | BUG/DEP: Make ufunclike functions more ufunc-like
| * | BUG: Return scalars from ufunclike objectsEric Wieser2017-04-262-17/+35
| | | | | | | | | | | | | | | | | | No need to reinvent the wheel here - the ufunc machinery will handle the out arguments Fixes #8993
| * | DEP: Deprecate calling the `out` argument `y`Eric Wieser2017-04-262-20/+59
| | | | | | | | | | | | Fixes gh-8995
* | | Merge pull request #8988 from eric-wieser/document-2522Charles Harris2017-04-271-0/+17
|\ \ \ | | | | | | | | DOC: Explain the behavior of diff on unsigned types
| * | | DOC: Explain the behavior of diff on unsigned typesEric Wieser2017-04-251-0/+17
| |/ / | | | | | | | | | Fixes #2522
* | | MAINT: Set the __name__ of generated methodsEric Wieser2017-04-271-30/+37
| |/ |/| | | | | Barely useful, but means that the function name is helpful if printed.
* | ENH: NDArrayOperatorsMixin calls ufuncs directly, like ndarrayStephan Hoyer2017-04-272-40/+36
| | | | | | | | | | | | | | | | | | * ENH: NDArrayOperatorsMixin calls ufuncs directly, like ndarray Per our discussion in https://github.com/numpy/numpy/pull/8247#discussion_r112825050 * add back in accidentally dropped __repr__
* | BUG: Fix ArrayLike(NDArrayOperatorsMixin) operations with object()Stephan Hoyer2017-04-272-12/+31
| |
* | ENH: Add NDArrayOperatorsMixin mixin class.Stephan Hoyer2017-04-273-0/+358
|/ | | | | | This mixin class provides an easy way to implement arithmetic operators that defer to __array_ufunc__ like numpy.ndarray in non-ndarray subclasses.
* DOC: Correct shape of edges in np.histogram2d (#8980)Importance of Being Ernest2017-04-231-2/+2
| | | | | If nx and ny are the bin counts (as stated in the bins argument's text), then the return H has indeed shape (nx, ny). However, the returned xedges and yedges will then have shape (nx+1,) and (ny+1,) respectively. Fixes #8979
* Merge pull request #6632 from mcmtroffaes/feature/fromfile-ioopen-bugCharles Harris2017-04-141-2/+3
|\ | | | | TST/BUG: fromfile - fix test and expose bug with io class argument
| * BUG: npy_PyFile_Dup2 - fix PyFile_AsFile failing on io style classesMatthias C. M. Troffaes2016-09-081-2/+3
| | | | | | | | | | | | | | | | | | | | This bug only manifests itself in the Python 2 code path. Falls back to the current Python 3 code path also on Python 2 because the Python 3 code path is written precisely to handle this situation. Also fix tests, and clarify in the documentation that the current implementation requires the stream to be seekable.
* | Merge pull request #8643 from eric-wieser/fix-8642Eric Wieser2017-04-112-1/+18
|\ \ | | | | | | BUG: Fix double-wrapping of object scalars
| * | BUG: Fix double-wrapping of object scalarsEric Wieser2017-04-062-1/+18
| | | | | | | | | | | | Fixes #8642
* | | DOC: Fix typos in percentile (#8900)Baurzhan Muftakhidinov2017-04-062-4/+4
| | |
* | | Merge pull request #8737 from eric-wieser/squeeze-expand-docsCharles Harris2017-04-011-0/+2
|\ \ \ | | | | | | | | DOC: Mention that expand_dims and squeeze are inverses
| * | | DOC: expand_dims and squeeze are inversesEric Wieser2017-03-061-0/+2
| | | | | | | | | | | | | | | | [ci skip]
* | | | MAINT: Rename _validate_axis, and document itEric Wieser2017-03-281-2/+2
| | | |
* | | | MAINT: Use _validate_axis inside _ureduceEric Wieser2017-03-283-33/+33
| | | | | | | | | | | | | | | | This fixes an omission where duplicate axes would only be detected when positive
* | | | MAINT: Reuse _validate_axis in np.gradientEric Wieser2017-03-282-16/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This also means that its axis argument invokes operator.index like others do. _validate_axis currently accepts lists of axes, which is a bug.
* | | | DOC: Spell out note for `dstack`Søren Fuglede Jørgensen2017-03-261-1/+2
| | | | | | | | | | | | This adds to the documentation on `dstack` the notes from `hstack` and `vstack` about dimensionality requirements.
* | | | Merge pull request #8388 from gfyoung/real-imag-scalarseberg2017-03-262-12/+66
|\ \ \ \ | | | | | | | | | | API: Return scalars for scalar inputs to np.real/imag
| * | | | API: Return scalars for scalar inputs to np.real/imaggfyoung2017-03-252-12/+66
| | | | | | | | | | | | | | | | | | | | Closes gh-8386.
* | | | | Merge pull request #8348 from anntzer/bincount-zero-minlengthEric Wieser2017-03-261-8/+7
|\ \ \ \ \ | |/ / / / |/| | | | ENH: Allow bincount(..., minlength=0).
| * | | | ENH: Allow bincount(..., minlength=0).Antony Lee2017-03-241-8/+7
| | | | |
* | | | | MAINT: Replace unicode() with u prefixEric Wieser2017-03-252-5/+2
| | | | |
* | | | | MAINT: Stop using sixu instead of a u prefixEric Wieser2017-03-251-2/+1
| | | | |
* | | | | MAINT: Remove asbytes_nested where b prefixes would sufficeEric Wieser2017-03-253-35/+33
| | | | |
* | | | | MAINT: Remove asbytes where a b prefix would sufficeEric Wieser2017-03-256-71/+70
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since we only need to support python 2, we can remove any case where we just pass a single string literal and use the b prefix instead. What we can't do is transform asbytes("tests %d" % num), because %-formatting fails on bytes in python 3.x < 3.5.
* | | | | Merge pull request #8813 from eric-wieser/maint-flatten-dtypeCharles Harris2017-03-232-5/+7
|\ \ \ \ \ | | | | | | | | | | | | MAINT: tidy up some of npyio
| * | | | | MAINT: Rename function to prevent name-shadowingEric Wieser2017-03-231-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This file had two functions called flatten_dtype, which did similar but different things.
| * | | | | DOC: There's a name for {False, True}, and it's "bool"Eric Wieser2017-03-231-2/+3
| | | | | |
View Lorry Controller Status