summaryrefslogtreecommitdiff
path: root/numpy
Commit message (Collapse)AuthorAgeFilesLines
* MAINT: random: Remove a few unused imports from test files.Warren Weckesser2019-12-153-3/+0
|
* Merge pull request #15069 from pv/blas-ilp64Charles Harris2019-12-1515-193/+230
|\ | | | | ENH: add support for ILP64 OpenBLAS (without symbol suffix)
| * BUG: core: use blas_ilp64 also for *_matmul, *_dot, and *_vdotPauli Virtanen2019-12-144-38/+53
| | | | | | | | Changing these to support ILP64 blas was missed in gh-15012
| * ENH: update BLAS symbol suffix/prefix handling in cblasfuncs & linalgPauli Virtanen2019-12-148-107/+45
| | | | | | | | Revise the BLAS name mangling to support the general scheme.
| * ENH: distutils: add support for ILP64 OpenBLAS (generic symbol suffix)Pauli Virtanen2019-12-143-48/+132
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Generalize the ILP64 BLAS/LAPACK symbol name handling to deal with arbitrary prefix/suffix. The build-time behavior is changed so that HAVE_BLAS_ILP64 and BLAS_SYMBOL_SUFFIX/PREFIX defines are added to compile options as appropriate. Mainly to make autodetection of BLAS/LAPACK easier for downstream numpy.distutils users, add get_info aliases 'blas_ilp64_opt', 'blas_ilp64_plain_opt', and 'blas64__opt' for any/no/""&"64_" prefix&suffix, and the same for lapack. (Due to the way system_info works, each also gets a separate class.) In addition to openblas64_ which has a fixed suffix, add the (by default suffixless) openblas_ilp64, which correspond to the most likely cases to be present.
* | TST: f2py: fix race condition in f2py test _get_compiler_statusPauli Virtanen2019-12-141-3/+9
| | | | | | | | | | | | Distutils code needs to run in different temporary directories, probably because they create `_configtest.c` files for compiler detection in the current directory.
* | ENH: f2py: add --f2cmap option for specifying the name of .f2py_f2cmapPauli Virtanen2019-12-144-16/+60
|/ | | | | | | | | | | | | Previously, f2py loaded the type mappings from a file ``.f2py_f2cmap`` in current directory, at import time. Make the file name customizable by adding a ``--f2cmap`` command line option, and postpone loading the file to f2py.run_main(). Moreover, restore the default type mapping in f2py.run_main() before loading the customizations, so that multiple calls to f2py.run_main() do not interfere with each other. (For example, numpy.distutils calls f2py multiple times in the same process.)
* DOC: bring the out parameter docstring into line with ufuncs (#15097)Matti Picus2019-12-141-7/+17
|
* ENH: Add support to sort timedelta64 `NaT` to end of the array (gh-15068)Tirth Patel2019-12-112-15/+35
| | | | | | This work is a follow up of gh-12658. As requested in gh-15063, add NaT sort support for timedelta64 datatypes also. Fixes gh-15063
* BUG: pickle the content of a scalar containing objects, not the address ↵Matti Picus2019-12-113-4/+51
| | | | | | | (gh-14940) Scalars with a void dtype that contains objects were not pickled properly. Add a test and fix by checking the NPY_LIST_PICKLE flag of the dtype. Fixes gh-13593
* Merge pull request #14981 from seberg/issue-13103Matti Picus2019-12-113-22/+31
|\ | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit c088383cb290ca064d456e89d79177a0e234cb8d and uses the same kind casting rule for the additional keyword arguments ``to_end`` and ``to_begin``. This results in slightly more leniant behaviour for integers (which can now have overflows that are hidden), but fixes an issue with the handling of NaN. Generally, this behaviour seems more conistent with what NumPy does elsewhere. The Overflow issue exists similar in many other places and should be solved by integer overflow warning machinery while the actual cast takes place. Closes gh-13103
| * BUG: Make ``ediff1d`` kwarg casting consistentSebastian Berg2019-11-253-22/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit c088383cb290ca064d456e89d79177a0e234cb8d and uses the same kind casting rule for the additional keyword arguments ``to_end`` and ``to_begin``. This results in slightly more leniant behaviour for integers (which can now have overflows that are hidden), but fixes an issue with the handling of NaN. Generally, this behaviour seems more conistent with what NumPy does elsewhere. The Overflow issue exists similar in many other places and should be solved by integer overflow warning machinery while the actual cast takes place. Closes gh-13103
* | Merge pull request #15035 from mattip/issue-14625Sebastian Berg2019-12-093-2/+43
|\ \ | | | | | | | | | Test, fix for missing end words in fortran subroutines and functions. Fixes gh-14625
| * | MAINT: fixes from reviewmattip2019-12-031-3/+4
| | |
| * | DOC: fix outstanding typo in NumPy versionmattip2019-12-031-1/+1
| | |
| * | BUG: add endfunction, endsubroutine to valid fortran end wordsmattip2019-12-032-1/+41
| | |
* | | Merge pull request #15072 from pv/requires-memory-fixCharles Harris2019-12-081-1/+24
|\ \ \ | | | | | | | | TST: testing: check requires_memory immediately before the test
| * | | TST: testing: check requires_memory immediately before the test + ignore ↵Pauli Virtanen2019-12-081-1/+24
| | | | | | | | | | | | | | | | MemoryErrors
* | | | MAINT: core: Fix spacing in the expm1 docstring.Warren Weckesser2019-12-081-1/+1
| | | |
* | | | MAINT: core: Fix a very long line in generated ufunc docstrings.Warren Weckesser2019-12-081-38/+74
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Put the $BROADCASTABLE_2 text on its own line, and add a line break in the text. This changes a typical ufunc docstring from this (note the very long line beginning 'The arrays to be subtracted...'): ``` Parameters ---------- x1, x2 : array_like The arrays to be subtracted from each other. If ``x1.shape != x2.shape``, they must be broadcastable to a common shape (which becomes the shape of the output). out : ndarray, None, or tuple of ndarray and None, optional A location into which the result is stored. If provided, it must have ``` to this: ``` Parameters ---------- x1, x2 : array_like The arrays to be subtracted from each other. If ``x1.shape != x2.shape``, they must be broadcastable to a common shape (which becomes the shape of the output). out : ndarray, None, or tuple of ndarray and None, optional A location into which the result is stored. If provided, it must have ```
* | | DOC: correct version for NaT sortTyler Reddy2019-12-071-1/+1
| | | | | | | | | | | | | | | | | | | | | * NaT sorting to the end of arrays was adjusted in NumPy 1.18.0 instead of 1.17.0, so adjust sort() docs accordingly
* | | DOC: clarify documentation for transpose() (gh-15024)Daniel Povey2019-12-061-8/+13
| | | | | | | | | Clarify axes argument and general text of `np.transpose`.
* | | Revert "DEP: issue deprecation warning when creating ragged array (NEP 34)"revert-14794-nep-0034-implRalf Gommers2019-12-0613-116/+51
| | |
* | | Merge pull request #15052 from pv/blas64-fixupSebastian Berg2019-12-065-12/+8
|\ \ \ | | | | | | | | MAINT: follow-up cleanup for blas64 PR
| * | | BUG: distutils: remove stray debug printPauli Virtanen2019-12-041-1/+0
| | | |
| * | | TST: fix up issues in requires_memory decoratorPauli Virtanen2019-12-044-11/+8
| | | | | | | | | | | | | | | | Fix wrong multiplier for /proc/meminfo, and do style cleanups.
* | | | DOC: Fix statement about norms (#15050)Andreas Klöckner2019-12-051-1/+1
| | | | | | | | | | | | The "0.5-norm" violates the triangle inequality because its unit ball is nonconvex.
* | | | Merge pull request #15022 from eric-wieser/better-f2py-errorMatti Picus2019-12-054-31/+14
|\ \ \ \ | | | | | | | | | | ENH: f2py: improve error messages
| * | | | ENH: Chain exceptions when converting python objects to fortranEric Wieser2019-12-023-28/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Note that since we now need `npy_3kcompat.h`, we can delete all duplicated lines in this file We preserve the original exception type for compatibility with old code here.
| * | | | ENH: Show the incorrect type in the fortran error messageEric Wieser2019-12-011-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I just got this error while hacking some scipy internals, and had nowhere near enough information to debug it. This is a little closer to useful.
* | | | | API, DOC: change names to multivariate_hypergeometric, improve docsmattip2019-12-044-14/+14
| |/ / / |/| | |
* | | | TST: Add test for object method (and general unary) loops (#15040)Sebastian Berg2019-12-041-1/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * TST: Add test for object method (and general unary) loops These loops should give the same results as when run on the equivalent numeric/floatingpoint values. Adds additional tests for gh-15036
* | | | REL: Update master after 1.18.x branch.Charles Harris2019-12-032-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Apart from the usual modifications, this - Removes a release note snippet added after towncrier was run - Removes testing on Python 3.5 as it is dropped for 1.19.x.
* | | | Merge pull request #15007 from mattip/random-c-api2Ralf Gommers2019-12-035-93/+96
|\ \ \ \ | | | | | | | | | | DOC, API: improve the C-API/Cython documentation and interfaces for random
| * | | | API: revert changes to standard_t, cauchymattip2019-11-303-8/+8
| | | | |
| * | | | API: rename functions in distributions.c,hmattip2019-11-293-66/+55
| | | | |
| * | | | DOC, TST: refactor CFFI test, add file names to documentationmattip2019-11-292-35/+49
| | | | |
* | | | | Merge pull request #15039 from eric-wieser/fix-weird-reraiseSebastian Berg2019-12-032-4/+2
|\ \ \ \ \ | | | | | | | | | | | | BUG: Exceptions tracebacks are dropped
| * | | | | BUG: Exceptions tracebacks are droppedEric Wieser2019-12-032-4/+2
| | |/ / / | |/| | | | | | | | | | | | | For some reason this code reconstructed brand new exception objects with no traceback, rather than just re-raising
* | | | | Merge pull request #15036 from eric-wieser/fix-leakMatti Picus2019-12-031-1/+5
|\ \ \ \ \ | | | | | | | | | | | | BUG: Fix refcounting in ufunc object loops
| * | | | | BUG: Fix refcounting in ufunc object loopsEric Wieser2019-12-031-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | Previously `func` was leaked on both success and failure paths.
* | | | | | Merge pull request #15023 from qwhelan/nan_perfMatti Picus2019-12-032-2/+30
|\ \ \ \ \ \ | | | | | | | | | | | | | | MAINT: Only copy input array in _replace_nan() if there are nans to replace
| * | | | | | PERF: only copy input array in _replace_nan() if there are nans to replaceChristopher Whelan2019-12-022-2/+30
| | |/ / / / | |/| | | |
* | | | | | MAINT: Cleaned up mintypecode for Py3Joseph R. Fox-Rabinovitz2019-12-021-1/+1
| |_|_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | Using generators instead of full-blown lists Using set for search instead of list Using min to get single element insteaf of sorting full list
* | | | | Merge pull request #14794 from mattip/nep-0034-implRalf Gommers2019-12-0213-51/+116
|\ \ \ \ \ | | | | | | | | | | | | DEP: issue deprecation warning when creating ragged array (NEP 34)
| * | | | | MAINT: changes from reviewmattip2019-11-291-2/+4
| | | | | |
| * | | | | MAINT: add test to test_deprecationsmattip2019-11-291-1/+9
| | | | | |
| * | | | | MAINT: fixes from reviewmattip2019-11-251-4/+3
| | | | | |
| * | | | | MAINT: fixes from reviewmattip2019-11-071-6/+7
| | | | | |
| * | | | | TST: fix failing testmattip2019-10-311-1/+1
| | | | | |
View Lorry Controller Status