Contributors

A total of 133 people contributed to this release. People with a "+" by their names contributed a patch for the first time.

• Aaron Critchley +
• Aarthi +
• Aarthi Agurusa +
• Alex Thomas +
• Alexander Belopolsky
• Allan Haldane
• Anas Khan +
• Andras Deak
• Andrey Portnoy +
• Anna Chiara
• Aurelien Jarno +
• Baurzhan Muftakhidinov
• Berend Kapelle +
• Bernhard M. Wiedemann
• Bjoern Thiel +
• Bob Eldering
• Cenny Wenner +
• Charles Harris
• ChloeColeongco +
• Chris Billington +
• Christopher +
• Chun-Wei Yuan +
• Claudio Freire +
• Daniel Smith
• Darcy Meyer +
• David Abdurachmanov +
• David Freese
• Deepak Kumar Gouda +
• Dennis Weyland +
• Derrick Williams +
• Dmitriy Shalyga +
• Eric Cousineau +
• Eric Larson
• Eric Wieser
• Evgeni Burovski
• Frederick Lefebvre +
• Gaspar Karm +
• Geoffrey Irving
• Gerhard Hobler +
• Gerrit Holl
• Guo Ci +
• Hameer Abbasi +
• Han Shen
• Hiroyuki V. Yamazaki +
• Hong Xu
• Ihor Melnyk +
• Jaime Fernandez
• Jake VanderPlas +
• James Tocknell +
• Jarrod Millman
• Jeff VanOss +
• John Kirkham
• Jonas Rauber +
• Jonathan March +
• Joseph Fox-Rabinovitz
• Julian Taylor
• Junjie Bai +
• Juris Bogusevs +
• Jörg Döpfert
• Kenichi Maehashi +
• Kevin Sheppard
• Kimikazu Kato +
• Kirit Thadaka +
• Kritika Jalan +
• Kyle Sunden +
• Lakshay Garg +
• Lars G +
• Licht Takeuchi
• Louis Potok +
• Luke Zoltan Kelley
• MSeifert04 +
• Mads R. B. Kristensen +
• Malcolm Smith +
• Mark Harfouche +
• Marten H. van Kerkwijk +
• Marten van Kerkwijk
• Matheus Vieira Portela +
• Mathieu Lamarre
• Mathieu Sornay +
• Matthew Brett
• Matthew Rocklin +
• Matthias Bussonnier
• Matti Picus
• Michael Droettboom
• Miguel Sánchez de León Peque +
• Mike Toews +
• Milo +
• Nathaniel J. Smith
• Nelle Varoquaux
• Nicholas Nadeau, P.Eng., AVS +
• Nick Minkyu Lee +
• Nikita +
• Nikita Kartashov +
• Nils Becker +
• Oleg Zabluda
• Orestis Floros +
• Pat Gunn +
• Paul van Mulbregt +
• Pauli Virtanen
• Pierre Chanial +
• Ralf Gommers
• Raunak Shah +
• Robert Kern
• Russell Keith-Magee +
• Ryan Soklaski +
• Samuel Jackson +
• Sebastian Berg
• Siavash Eliasi +
• Simon Conseil
• Simon Gibbons
• Stefan Krah +
• Stefan van der Walt
• Stephan Hoyer
• Subhendu +
• Subhendu Ranjan Mishra +
• Tai-Lin Wu +
• Tobias Fischer +
• Toshiki Kataoka +
• Tyler Reddy +
• Unknown +
• Varun Nayyar
• Victor Rodriguez +
• Warren Weckesser
• William D. Irons +
• Zane Bradley +
• cclauss +
• fo40225 +
• lapack_lite code generator +
• lumbric +
• luzpaz +
• mamrehn +
• tynn +
• xoviat

Pull requests merged

A total of 438 pull requests were merged for this release.

• #8157: BUG: void .item() doesn't hold reference to original array
• #8774: ENH: Add gcd and lcm ufuncs
• #8819: ENH: Implement axes keyword argument for gufuncs.
• #8952: MAINT: Removed duplicated code around ufunc->identity
• #9686: DEP: Deprecate non-tuple nd-indices
• #9980: MAINT: Implement lstsq as a gufunc
• #9998: ENH: Nditer as context manager
• #10073: ENH: Implement fft.fftshift/ifftshift with np.roll for improved...
• #10078: DOC: document nested_iters
• #10128: BUG: Prefix library names with lib on windows.
• #10142: DEP: Pending deprecation warning for matrix
• #10154: MAINT: Use a StructSequence in place of the typeinfo tuples
• #10158: BUG: Fix a few smaller valgrind errors
• #10178: MAINT: Prepare master for 1.15 development.
• #10186: MAINT: Move histogram and histogramdd into their own module
• #10187: BUG: Extra space is inserted on first line for long elements
• #10192: DEP: Deprecate the pickle aliases
• #10193: BUG: Fix bugs found by testing in release mode.
• #10194: BUG, MAINT: Ufunc reduce reference leak
• #10195: DOC: Fixup percentile docstring, from review in gh-9213
• #10196: BUG: Fix regression in np.ma.load in gh-10055
• #10199: ENH: Quantile
• #10203: MAINT: Update development branch version to 1.15.0.
• #10205: BUG: Handle NaNs correctly in arange
• #10207: ENH: Allow np.r_ to accept 0d arrays
• #10208: MAINT: Improve error message for void(-1)
• #10210: DOC: change 'a'->'prototype' in empty_like docs (addresses #10209)
• #10211: MAINT,ENH: remove MaskedArray.astype, as the base type does everything.
• #10212: DOC: fix minor typos
• #10213: ENH: Set up proposed NEP process
• #10214: DOC: add warning to isclose function
• #10216: BUG: Fix broken format string picked up by LGTM.com
• #10220: DOC: clarify that np.absolute == np.abs
• #10223: ENH: added masked version of 'numpy.stack' with tests.
• #10225: ENH: distutils: parallelize builds by default
• #10226: BUG: distutils: use correct top-level package name
• #10229: BUG: distutils: fix extra DLL loading in certain scenarios
• #10231: BUG: Fix sign-compare warnings in datetime.c and datetime_strings.c.
• #10232: BUG: Don't reimplement isclose in np.ma
• #10237: DOC: give correct version of np.nansum change
• #10241: MAINT: Avoid repeated validation of percentiles in nanpercentile
• #10247: MAINT: fix typo
• #10248: DOC: Add installation notes for Linux users
• #10249: MAINT: Fix tests failures on travis CI merge.
• #10250: MAINT: Check for __array_ufunc__ before doing anything else.
• #10251: ENH: Enable AVX2/AVX512 support to numpy
• #10252: MAINT: Workaround for new travis sdist failures.
• #10255: MAINT: Fix loop and simd sign-compare warnings.
• #10257: BUG: duplicate message print if warning raises an exception
• #10259: BUG: Make sure einsum default value of optimize is True.
• #10260: ENH: Add pytest support
• #10261: MAINT: Extract helper functions from histogram
• #10262: DOC: Add missing release note for #10207
• #10263: BUG: Fix strange behavior of infinite-step-size/underflow-case...
• #10264: MAINT: Fix (some) yield warnings
• #10266: BUG: distutils: fix locale decoding errors
• #10268: BUG: Fix misleading error when coercing to array
• #10269: MAINT: extract private helper function to compute histogram bin...
• #10271: BUG: Allow nan values in the data when the bins are explicit
• #10278: ENH: Add support for datetimes to histograms
• #10282: MAINT: Extract helper function for last-bound-inclusive search_sorted
• #10283: MAINT: Fallback on the default sequence multiplication behavior
• #10284: MAINT/BUG: Tidy gen_umath
• #10286: BUG: Fix memory leak (#10157).
• #10287: ENH: Allow ptp to take an axis tuple and keepdims
• #10292: BUG: Masked singleton can be reshaped to be non-scalar
• #10293: MAINT: Fix sign-compare warnings in mem_overlap.c.
• #10294: MAINT: pytest cleanups
• #10298: DOC: Explain np.digitize and np.searchsorted more clearly
• #10300: MAINT, DOC: Documentation and misc. typos
• #10303: MAINT: Array wrap/prepare identification cleanup
• #10309: MAINT: deduplicate check_nonreorderable_axes
• #10314: BUG: Ensure __array_finalize__ cannot back-mangle shape
• #10316: DOC: add documentation about how to handle new array printing
• #10320: BUG: skip the extra-dll directory when there are no DLLS
• #10323: MAINT: Remove duplicated code for promoting dtype and array types.
• #10324: BUG: Fix crashes when using float32 values in uniform histograms
• #10325: MAINT: Replace manual expansion of PyArray_MinScalarType with...
• #10327: MAINT: Fix misc. typos
• #10333: DOC: typo fix in numpy.linalg.det docstring
• #10334: DOC: Fix typos in docs for partition method
• #10336: DOC: Post 1.14.0 release updates.
• #10337: ENH: Show the silenced error and traceback in warning __cause__
• #10341: BUG: fix config where PATH isn't set on win32
• #10342: BUG: arrays not being flattened in union1d
• #10346: ENH: Check matching inputs/outputs in umath generation
• #10352: BUG: Fix einsum optimize logic for singleton dimensions
• #10354: BUG: fix error message not formatted in einsum
• #10359: BUG: do not optimize einsum with only 2 arguments.
• #10361: BUG: complex repr has extra spaces, missing +
• #10362: MAINT: Update download URL in setup.py.
• #10367: BUG: add missing paren and remove quotes from repr of fieldless...
• #10371: BUG: fix einsum issue with unicode input and py2
• #10381: BUG/ENH: Improve output for structured non-void types
• #10388: ENH: Add types for int and uint of explicit sizes to swig.
• #10390: MAINT: Adjust type promotion in linalg.norm
• #10391: BUG: Make dtype.descr error for out-of-order fields
• #10392: DOC: Document behaviour of np.concatenate with axis=None
• #10401: BUG: Resize bytes_ columns in genfromtxt
• #10402: DOC: added "steals a reference" to PyArray_FromAny
• #10406: ENH: add np.printoptions, a context manager
• #10411: BUG: Revert multifield-indexing adds padding bytes for NumPy...
• #10412: ENH: Fix repr of np.record objects to match np.void types
• #10414: MAINT: Fix sign-compare warnings in umath_linalg.
• #10415: MAINT: Fix sign-compare warnings in npy_binsearch, npy_partition.
• #10416: MAINT: Fix sign-compare warnings in dragon4.c.
• #10418: MAINT: Remove repeated #ifdefs implementing isinstance(x, basestring)...
• #10420: DOC: Fix version added labels in numpy.unique docs
• #10421: DOC: Fix type of axis in nanfunctions
• #10423: MAINT: Update zesty to artful for i386 testing
• #10426: DOC: Add version when linalg.norm accepted axis
• #10427: DOC: Fix typo in docs for argpartition
• #10430: MAINT: Use ValueError for duplicate field names in lookup
• #10433: DOC: Add 1.14.1 release notes template (forward port)
• #10434: MAINT: Move tools/announce.py to tools/changelog.py.
• #10441: BUG: Fix nan_to_num return with integer input
• #10443: BUG: Fix various Big-Endian test failures (ppc64)
• #10444: MAINT: Implement float128 dragon4 for IBM double-double (ppc64)
• #10451: BUG: prevent the MSVC 14.1 compiler (Visual Studio 2017) from...
• #10453: Revert "BUG: prevent the MSVC 14.1 compiler (Visual Studio 2017)...
• #10458: BLD: Use zip_safe=False in setup() call
• #10459: MAINT: Remove duplicated logic between array_wrap and array_prepare
• #10463: ENH: Add entry_points for f2py, conv_template, and from_template.
• #10465: MAINT: Fix miscellaneous sign-compare warnings.
• #10472: DOC: Document A@B in Matlab/NumPy summary table
• #10473: BUG: Fixed polydiv for Complex Numbers
• #10475: DOC: Add CircleCI builder for devdocs
• #10476: DOC: fix formatting in interp example
• #10477: BUG: Align type definition with generated lapack
• #10478: DOC: Minor punctuation cleanups and improved explanation.
• #10479: BUG: Fix calling ufuncs with a positional output argument.
• #10482: BUG: Add missing DECREF in Py2 int() cast
• #10484: MAINT: Remove unused code path for applying maskedarray domains...
• #10497: DOC: Tell matlab users about np.block
• #10498: MAINT: Remove special cases in np.unique
• #10501: BUG: fromregex: asbytes called on regexp objects
• #10502: MAINT: Use AxisError in swapaxes, unique, and diagonal
• #10503: BUG: Fix unused-result warning.
• #10506: MAINT: Delete unused _build_utils/common.py
• #10508: BUG: Add missing #define _MULTIARRAYMODULE to vdot.c
• #10509: MAINT: Use new-style format strings for clarity
• #10516: MAINT: Allow errors to escape from InitOperators
• #10518: ENH: Add a repr to np._NoValue
• #10522: MAINT: Remove the unmaintained umath __version__ constant.
• #10524: BUG: fix np.save issue with python 2.7.5
• #10529: BUG: Provide a better error message for out-of-order fields
• #10543: DEP: Issue FutureWarning when malformed records detected.
• #10544: BUG: infinite recursion in str of 0d subclasses
• #10546: BUG: In numpy.i, clear CARRAY flag if wrapped buffer is not C_CONTIGUOUS.
• #10547: DOC: Fix incorrect formula in gradient docstring.
• #10548: BUG: Set missing exception after malloc
• #10549: ENH: Make NpzFile conform to the Mapping protocol
• #10553: MAINT: Cleanups to promote_types and result_types
• #10554: DOC: promote_types is not associative by design,
• #10555: BUG: Add missing PyErr_NoMemory() after malloc
• #10564: BUG: Provide correct format in Py_buffer for scalars
• #10566: BUG: Fix travis failure in previous commit
• #10571: BUG: Fix corner-case behavior of cond() and use SVD when possible
• #10576: MAINT: Fix misc. documentation typos
• #10583: MAINT: Fix typos in DISTUTILS.rst.
• #10588: BUG: Revert sort optimization in np.unique.
• #10589: BUG: fix entry_points typo for from-template
• #10591: ENH: Add histogram_bin_edges function and test
• #10592: DOC: Corrected url for Guide to NumPy book; see part of #8520,...
• #10596: MAINT: Update sphinxext submodule hash.
• #10599: ENH: Make flatnonzero call asanyarray before ravel()
• #10603: MAINT: Improve error message in histogram.
• #10604: MAINT: Fix Misc. typos
• #10606: MAINT: Do not use random roots when testing roots.
• #10618: MAINT: Stop using non-tuple indices internally
• #10619: BUG: np.ma.flatnotmasked_contiguous behaves differently on mask=nomask...
• #10621: BUG: deallocate recursive closure in arrayprint.py
• #10623: BUG: Correctly identify comma separated dtype strings
• #10625: BUG: Improve the accuracy of the FFT implementation
• #10635: ENH: Implement initial kwarg for ufunc.add.reduce
• #10641: MAINT: Post 1.14.1 release updates for master branch
• #10650: BUG: Fix missing NPY_VISIBILITY_HIDDEN on npy_longdouble_to_PyLong
• #10653: MAINT: Remove duplicate implementation for aliased functions.
• #10657: BUG: f2py: fix f2py generated code to work on Pypy
• #10658: BUG: Make np.partition and np.sort work on np.matrix when axis=None
• #10660: BUG/MAINT: Remove special cases for 0d arrays in interp
• #10661: MAINT: Unify reductions in fromnumeric.py
• #10665: ENH: umath: don't make temporary copies for in-place accumulation
• #10666: BUG: fix complex casting error in cov with aweights
• #10669: MAINT: Covariance must be symmetric as well as positive-semidefinite.
• #10670: DEP: Deprecate np.sum(generator)
• #10671: DOC/MAINT: More misc. typos
• #10672: ENH: Allow dtype field names to be ascii encoded unicode in Python2
• #10676: BUG: F2py mishandles quoted control characters
• #10677: STY: Minor stylistic cleanup of numeric.py
• #10679: DOC: zeros, empty, and ones now have consistent docstrings
• #10684: ENH: Modify intersect1d to return common indices
• #10689: BLD: Add configuration changes to allow cross platform builds...
• #10691: DOC: add versionadded for NDArrayOperatorsMixin.
• #10694: DOC: Improve docstring of memmap
• #10698: BUG: Further back-compat fix for subclassed array repr (forward...
• #10699: DOC: Grammar of np.gradient docstring
• #10702: TST, DOC: Upload devdocs and neps after circleci build
• #10703: MAINT: NEP process updates
• #10708: BUG: fix problem with modifying pyf lines containing ';' in f2py
• #10710: BUG: fix error message in numpy.select
• #10711: MAINT: Hard tab and whitespace cleanup.
• #10715: MAINT: Fixed C++ guard in f2py test.
• #10716: BUG: dragon4 fractional output mode adds too many trailing zeros
• #10718: BUG: Fix bug in asserting near equality of float16 arrays.
• #10719: DOC: add documentation for constants
• #10720: BUG: distutils: Remove named templates from the processed output...
• #10722: MAINT: Misc small fixes.
• #10730: DOC: Fix minor typo in how-to-document.
• #10732: BUG: Fix setup.py build install egg_info, which did not previously...
• #10734: DOC: Post 1.14.2 release update.
• #10737: MAINT: Fix low-hanging PyPy compatibility issues
• #10739: BUG: Fix histogram bins="auto" for data with little variance
• #10740: MAINT, TST: Fixes for Python 3.7
• #10743: MAINT: Import abstract classes from collections.abc
• #10745: ENH: Add object loops to the comparison ufuncs
• #10746: MAINT: Fix typo in warning message
• #10748: DOC: a.size and np.prod(a.shape) are not equivalent
• #10750: DOC: Add graph showing different behaviors of np.percentile
• #10755: DOC: Move bin estimator documentation from histogram to histogram_bin_edges
• #10758: TST: Change most travisci tests to Python3.6.
• #10763: BUG: floating types should override tp_print
• #10766: MAINT: Remove the unused scalarmath getters for fmod and sqrt
• #10773: BUG: Use dummy_threading on platforms that don't support threading
• #10774: BUG: Fix SQRT_MIN for platforms with 8-byte long double
• #10775: BUG: Return NULL from PyInit_* when exception is raised
• #10777: MAINT: Remove use of unittest in NumPy tests.
• #10778: BUG: test, fix for missing flags['WRITEBACKIFCOPY'] key
• #10781: ENH: NEP index builder
• #10785: DOC: Fixed author name in reference to book
• #10786: ENH: Add "stable" option to np.sort as an alias for "mergesort".
• #10790: TST: Various fixes prior to switching to pytest
• #10795: BUG: Allow spaces in output string of einsum
• #10796: BUG: fix wrong inplace vectorization on overlapping arguments
• #10798: BUG: error checking before mapping of einsum axes.
• #10800: DOC: Add remarks about array vs scalar output to every ufunc
• #10802: BUG/DOC/MAINT: Tidy up histogramdd
• #10807: DOC: Update link to tox in development docs (#10806)
• #10812: MAINT: Rearrange numpy/testing files
• #10814: BUG: verify the OS supports avx instruction
• #10822: BUG: fixes exception in numpy.genfromtxt, see #10780
• #10824: BUG: test, fix PyArray_DiscardWritebackIfCopy refcount issue...
• #10826: BUG: np.squeeze() now respects older API axis expectation
• #10827: ENH: Add tester for pytest.
• #10828: BUG: fix obvious mistake in testing/decorators warning.
• #10829: BLD: use Python 3.6 instead of 2.7 as default for doc build.
• #10830: BUG: Fix obvious warning bugs.
• #10831: DOC: Fix minor typos
• #10832: ENH: datetime64: support AC dates starting with '+'
• #10833: ENH: Add support for the 64-bit RISC-V architecture
• #10834: DOC: note that NDEBUG should be set when OPT should increase...
• #10836: MAINT: Fix script name for pushing NEP docs to repo
• #10840: MAINT: Fix typo in code example.
• #10842: TST: Switch to pytest
• #10849: DOC: fix examples in docstring for np.flip
• #10850: DEP: Issue deprecation warnings for some imports.
• #10858: MAINT: Post pytest switch cleanup
• #10859: MAINT: Remove yield tests
• #10860: BUG: core: fix NPY_TITLE_KEY macro on pypy
• #10863: MAINT: More Histogramdd cleanup
• #10867: DOC: Cross Link full/full_like in a few see-also sections.
• #10869: BUG: Fix encoding regression in ma/bench.py (Issue #10868)
• #10871: MAINT: Remove unnecessary special case in np.histogramdd for...
• #10872: ENH: Extend np.flip to work over multiple axes
• #10874: DOC: State in docstring that lexsort is stable (#10873).
• #10875: BUG: fix savetxt, loadtxt for '+-' in complex
• #10878: DOC: rework documents and silence warnings during sphinx build
• #10882: BUG: have _array_from_buffer_3118 correctly handle errors
• #10883: DOC: Fix negative binomial documentation.
• #10885: TST: Re-enable test display on appveyor
• #10890: MAINT: lstsq: compute residuals inside the ufunc
• #10891: TST: Extract a helper function to test for reference cycles
• #10898: ENH: Have dtype transfer for equivalent user dtypes prefer user-defined...
• #10901: DOC, BUG : Bad link to np.random.randint
• #10903: DOC: Fix link in See Also section of randn docstring.
• #10907: TST: reactivate module docstring tests, fix float formatting
• #10911: BUG: Fix casting between npy_half and float in einsum
• #10916: BUG: Add missing underscore to prototype in check_embedded_lapack
• #10919: BUG: Pass non-None outputs to __array_prepare__ and __array_wrap__
• #10921: DOC: clear up warnings, fix matplotlib plot
• #10923: BUG: fixed dtype alignment for array of structs in case of converting...
• #10925: DOC: Fix typos in 1.15.0 changelog
• #10936: DOC: Fix NumpyVersion example (closes gh-10935)
• #10938: MAINT: One step closer to vectorizing lstsq
• #10940: DOC: fix broken links for developer documentation
• #10943: ENH: Add a search box to the sidebar in the docs
• #10945: MAINT: Remove references to the 2008 documentation marathon
• #10946: BUG: 'style' arg to array2string broken in legacy mode
• #10949: DOC: cleanup documentation, continuation of nditer PR #9998
• #10951: BUG: it.close() disallows access to iterator, fixes #10950
• #10953: MAINT: address extraneous shape tuple checks in descriptor.c
• #10958: MAINT, DOC: Fix typos
• #10967: DOC: add quantile, nanquantile to toc
• #10970: WIP: Remove fragile use of __array_interface__ in ctypeslib.as_array
• #10971: MAINT: Remove workaround for gh-10891
• #10973: DOC: advise against use of matrix.
• #10975: MAINT: move linalg tests using matrix to matrixlib
• #10980: DOC: link to governance, convert external link to internal
• #10984: MAINT: Added pytest cache folder to .gitignore
• #10985: MAINT, ENH: Move matrix_power to linalg and allow higher dimensions.
• #10986: MAINT: move all masked array matrix tests to matrixlib.
• #10987: DOC: Correction to docstring example (result was correct)
• #10988: MAINT: Small tidy-ups to ufunc_object.c
• #10991: DOC: Update genfromtxt docs to use StringIO and u-strings
• #10996: DOC: Make doc examples using StringIO python2-3 compatible
• #11003: DOC: work around GH isaacs/github#316 to show SVG image
• #11005: MAINT: Misc. typos
• #11006: TST, BUILD: add latex to circleci doc build
• #11008: REL: Fwd port 1.14.3 changelog
• #11009: DOC: release walkthrough updates from 1.14.3
• #11010: Move remaining Matrix tests to matrixlib
• #11011: MAINT: Simplify dimension-juggling in np.pad
• #11012: MAINT: np.pad: Add helper functions for producing slices along...
• #11018: ENH: Implement axis for generalized ufuncs.
• #11023: BUG: np.histogramdd loses precision on its inputs, leading to...
• #11026: MAINT: reduce code duplication in ufunc_frompyfunc
• #11033: BUG: Fix padding with large integers
• #11036: BUG: optimizing compilers can reorder call to npy_get_floatstatus
• #11037: BUG: initialize value before use
• #11038: ENH: Add __deepcopy__ to MaskedConstant
• #11043: BUG: reduce using SSE only warns if inside SSE loop
• #11050: BUG: remove fast scalar power for arrays with object dtype
• #11053: DOC: bump scipy-sphinx-theme to current version
• #11055: DOC: Add explanation for comments=None in loadtxt.
• #11056: MAINT: Improve performance of random permutation
• #11057: BUG: use absolute imports in test files
• #11066: MAINT: distutils.system_info: handle Accelerate like any other...
• #11073: DOC: expand reasoning behind npy_*floatstatus_barrer()
• #11076: BUG: Ensure PyArray_AssignRawScalar respects NPY_NEEDS_INIT
• #11082: DOC: link to updated module docstring, not NEP
• #11083: ENH: remove nose from travis tests
• #11085: DOC: create label and ref, fixes broken link
• #11086: DOC: Mention we can return unitinitialized values
• #11089: BLD: cleanup _configtest.o.d during build
• #11090: BUG: Added support for index values 27-52 in C einsum
• #11091: BUG: Python2 doubles don't print correctly in interactive shell
• #11094: DOC: add numpy.lib.format to docs and link to it
• #11095: MAINT: Einsum argument parsing cleanup
• #11097: BUG: fix datetime.timedelta->timedelta64 unit detection logic
• #11098: ENH: Add keepdims argument for generalized ufuncs.
• #11105: ENH: Add (put|take)_along_axis
• #11111: BUG: fix case of ISA selector in ufunc selection
• #11116: BUG: Typo in variable name in binary_repr
• #11120: MAINT: remove redundant code in MaskedArray.__new__
• #11122: BUG,MAINT: Ensure masked elements can be tested against nan and...
• #11124: BUG: Ensure that fully masked arrays pass assert_array_equal.
• #11134: DOC: Clarify tofile requirements
• #11137: MAINT: move remaining MaskedArray matrix tests to matrixlib.
• #11139: TST: turn some build warnings into errors
• #11140: MAINT: Update artful to bionic for i386 testing
• #11141: MAINT: Extract a helper function for prepending and appending
• #11145: DOC: cleanup NEP creation
• #11146: DOC: add a NEP to split MaskedArray into a separate package
• #11148: TST: make build warning into an error in runtest.py
• #11149: BUG: guessing datetime, time precedence
• #11152: BENCH: Add basic benchmarks for numpy.pad
• #11155: BUG: Prevent stackoverflow in conversion to datetime types
• #11158: TST: disable gc in refcount test
• #11159: TST: Skip ctypes dependent test that fails on Python < 2.7.7.
• #11160: TST: windows builds now properly support floating error states
• #11163: MAINT: Work around non-deterministic Python readdir order in...
• #11167: MAINT: Cleanup dragon4 code in various ways
• #11168: TST: linalg: add regression test for gh-8577
• #11169: MAINT: add sanity-checks to be run at import time
• #11173: MAINT: Ensure that parsing errors are passed on even in tests.
• #11176: MAINT: avoid setting non-existing gufunc strides for keepdims=True.
• #11177: DOC: improvement of the documentation for gufunc.
• #11178: TST: Test dimensions/indices found from parsed gufunc signatures.
• #11180: BUG: void dtype setup checked offset not actual pointer for alignment
• #11182: BUG: Avoid deprecated non-tuple indexing
• #11184: MAINT: Add bitmask helper functions
• #11185: MAINT: Add comments to long_double detection code
• #11186: TST: Add np.core._multiarray_tests.format_float_OSprintf_g
• #11187: MAINT: Use the more common -1 / 0 to indicate error / success
• #11189: NEP: Array function protocol
• #11190: DOC: Update NEP0 to clarify that discussion should happen on...
• #11191: MAINT: remove darwin hardcoded LDOUBLE detection
• #11193: BUG: Fix reference count/memory leak exposed by better testing
• #11200: BUG: Bytes delimiter/comments in genfromtxt should be decoded
• #11209: DOC: Fix doctest formatting in rot90() examples
• #11218: BUG: Fixes einsum broadcasting bug when optimize=True
• #11222: DOC: Make reference doc nditer examples python3 friendly
• #11223: BUG: Forcibly promote shape to uint64 in numpy.memmap.
• #11225: DOC: add existing recfunctions documentation to output
• #11226: MAINT: add 'rst' to nep filename, fixup urls
• #11229: NEP: New RNG policy
• #11231: MAINT: ensure we do not create unnecessary tuples for outputs
• #11238: MAINT: Don't update the flags a second time
• #11239: MAINT: Use PyArray_NewFromDescr where possible, remove unused...
• #11240: MAINT: Remove dead code backporting py2.6 warnings
• #11246: BUG: Set ndarray.base before __array_finalize__
• #11247: MAINT/BUG: Remove out-of-band reference count in PyArray_Newshape,...
• #11248: MAINT: Don't update the flags a second time
• #11249: BUG: Remove errant flag meddling in .real and .imag
• #11252: DOC: show how to generate release notes in release walkthrough
• #11257: BUG: ensure extobj and axes have their own references.
• #11260: MAINT: Do proper cleanup in get_ufunc_arguments.
• #11263: DOC: Update master after NumPy 1.14.4 release.
• #11269: BUG: Correct use of NPY_UNUSED.
• #11273: BUG: Remove invalid read in searchsorted if needle is empty
• #11275: TST: Do not use empty arrays in tests (unless they are not read)
• #11277: BUG: Work around past and present PEP3118 issues in ctypes
• #11280: DOC: make docstring of np.interp clearer
• #11286: BUG: einsum needs to check overlap on an out argument
• #11287: DOC: Minor documentation improvements
• #11291: BUG: Remove extra trailing parentheses.
• #11293: DOC: fix hierarchy of numericaltype
• #11296: BUG: Fix segfault on failing __array_wrap__
• #11298: BUG: Undo behavior change in ma.masked_values(shrink=True)
• #11307: BUG: Fix memmap regression when shape=None
• #11314: MAINT: remove unused "npy_import"
• #11315: MAINT: Package tools/allocation_tracking
• #11319: REL, REV: Revert f2py fixes that exposed SciPy bug.
• #11327: DOC: Update release notes for 1.15.0.
• #11339: BUG: decref in failure path; replace PyObject_Type by Py_TYPE
• #11352: DEP: Actually deprecate the normed argument to histogram
• #11359: DOC: document new functions
• #11367: BUG: add missing NpyIter_Close in einsum
• #11368: BUG/TST: String indexing should just fail, not emit a futurewarning
• #11389: ENH: Remove NpyIter_Close
• #11392: BUG: Make scalar.squeeze accept axis arg
• #11393: REL,MAINT: Update numpyconfig.h for 1.15.
• #11394: MAINT: Update mailmap
• #11403: DOC: Remove npyiter close from notes
• #11427: BUG: Fix incorrect deprecation logic for histogram(normed=...)...
• #11489: BUG: Ensure out is returned in einsum.
• #11491: BUG/ENH: Einsum optimization path updates and bug fixes.
• #11493: BUG: Revert #10229 to fix DLL loads on Windows.
• #11494: MAINT: add PyPI classifier for Python 3.7
• #11495: BENCH: belated addition of lcm, gcd to ufunc benchmark.
• #11496: BUG: Advanced indexing assignment incorrectly took 1-D fastpath
• #11511: BUG: Fix #define for ppc64 and ppc64le
• #11529: ENH: Add density argument to histogramdd.
• #11532: BUG: Decref of field title caused segfault
• #11540: DOC: Update the 1.15.0 release notes.
• #11577: BLD: Modify cpu detection and printing to get working aarch64...
• #11578: DOC: link to TESTS.rst testing guidelines, tweak testing...
• #11602: TST: Add Python 3.7 to CI testing