summaryrefslogtreecommitdiff
path: root/doc
Commit message (Collapse)AuthorAgeFilesLines
* Merge pull request #9842 from bashtage/protect-empty-initCharles Harris2017-10-181-0/+6
|\ | | | | BUG: Prevent invalid array shapes in seed
| * BUG: Prevent invalid array shapes in seedKevin Sheppard2017-10-181-0/+6
| | | | | | | | | | | | | | Prevent empty arrays or arrays with more than 1 dimension from being used to seed RandomState closes #9832
* | DOC: Refine SVD documentation (#9845)Toon Verstraelen2017-10-181-0/+2
| |
* | Merge pull request #9065 from eric-wieser/deprecate-bincountCharles Harris2017-10-181-0/+2
|\ \ | | | | | | DEP: 0 should be passed to bincount, not None
| * | DEP: 0 should be passed to bincount, not NoneEric Wieser2017-10-171-0/+2
| |/ | | | | | | Fixes gh-8841
* | BUG: count_nonzero treats empty axis tuples strangelyEric Wieser2017-10-171-0/+7
|/ | | | | | Fixes #9728 This bug was introduced with the `axis` keyword in #7177, as a misguided optimization.
* ENH: Save to ZIP files without using temporary files.Serhiy Storchaka2017-10-151-0/+5
| | | | | Since Python 3.6 it is possible to write directly to a ZIP file, without creating temporary files.
* Merge pull request #9835 from kenogo/masterCharles Harris2017-10-141-0/+6
|\ | | | | BENCH: Added missing ufunc benchmarks
| * BUG: Raise TypeError instead of ValueError for ufunc isnat if operand is not ↵Keno Goertz2017-10-111-0/+6
| | | | | | | | | | | | | | of type datetime or timedelta TST: Edited test_datetime to expect a TypeError instead of a ValueError for isnat if a wrong type is passed as an argument DOC: Added change of ufunc isnat's exception handling to release notes
* | Merge pull request #9856 from eric-wieser/fix-void-bool-2Charles Harris2017-10-141-0/+8
|\ \ | | | | | | BUG: Make bool(void_scalar) and void_scalar.astype(bool) consistent
| * | BUG: Make bool(void_scalar) and void_scalar.astype(bool) consistentEric Wieser2017-10-131-0/+8
| | | | | | | | | | | | | | | | | | Caused by void scalars decaying to 1d uint8 arrays before casting - `getitem` is dangerous for intermediate results Works towards #9847
* | | DOC: Some minor fixes regarding import_arrayMichael Seifert2017-10-132-5/+5
|/ / | | | | | | [skip ci]
* | Merge pull request #9718 from hemildesai/deprecate_truth_testing_on_empty_arraysEric Wieser2017-10-101-0/+2
|\ \ | |/ |/| DEP: Deprecate truth testing on empty arrays
| * DEP: Deprecate truth testing on empty arrayshemildesai2017-09-281-0/+2
| | | | | | | | Fixes #9583
* | DOC: update other MATLAB ode recommendationxoviat2017-10-021-2/+2
| |
* | DOC: update scipy.integrate recommendationxoviat2017-10-021-1/+1
| | | | | | Closes scipy/scipy#7965.
* | DOC: Add release notes for 0-size dtypesEric Wieser2017-10-011-0/+17
| |
* | DOC: add missing underscore in set_printoptionsLoïc Estève2017-10-011-1/+1
|/
* Merge pull request #9784 from eric-wieser/fix-void-recursionAllan Haldane2017-09-281-5/+12
|\ | | | | BUG: remove voidtype-repr recursion in scalartypes.c/arrayprint.py
| * BUG: remove voidtype-repr recursion in scalartypes.c/arrayprint.pyAllan Haldane2017-09-261-5/+12
| | | | | | | | Fixes #9345
* | DOC: Update after NumPy 1.13.2 release.Charles Harris2017-09-273-0/+106
|/ | | | | | | | * Add 1.13.2-changelog.rst * Add doc/release/1.13.2-notes.rst * Update doc/source/release.rst [ci skip]
* MAINT: fixups to legacy spacing in FloatFormatAllan Haldane2017-09-251-15/+5
|
* ENH: Simplify some code in arrayprint.pyAllan Haldane2017-09-251-9/+15
|
* ENH: "pad_sign" option, also threshold and edgeitemsAllan Haldane2017-09-251-6/+23
|
* ENH: remove unneeded spaces in float/bool reprsAllan Haldane2017-09-251-0/+7
|
* Merge pull request #9683 from MSeifert04/doc_add_polyutilsCharles Harris2017-09-232-0/+5
|\ | | | | DOC: Add polyutils subpackage to reference documentation
| * DOC: Add polyutils subpackage to reference documentation [skip ci]Michael Seifert2017-09-232-0/+5
| |
* | Merge pull request #9336 from eric-wieser/masked-constantCharles Harris2017-09-211-0/+10
|\ \ | | | | | | BUG: Fix various problems with the np.ma.masked constant
| * | DOC: Add release notes warning about np.ma.masked immutabilityEric Wieser2017-09-131-0/+10
| | |
* | | Fixing typo "Porland" -> "Portland" in `building` doc.Danny Hermes2017-09-191-2/+2
| | |
* | | Merge branch 'master' into concatenate-outEric Wieser2017-09-173-18/+87
|\ \ \
| * | | ENH: add hermitian=False kwarg to matrix_powerCJ Carey2017-09-161-0/+6
| | | | | | | | | | | | | | | | | | | | With a symmetric matrix, the more efficient `eigvalsh` method can be used to find singular values.
| * | | Merge branch 'master' into fix-pinvCharles Harris2017-09-132-2/+21
| |\ \ \
| | * | | ENH: Align data in np.save() at 64 bytes (#9025)John Zwinck2017-09-131-0/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously, saving format version 1 would align to 16 bytes, and saving version 2 would align improperly (bug #8085). Alignment is now always at least 64 bytes in either version, which supports memory mapping of the saved files on Linux, where mmap() offset must be a multiple of the page size. Why 64 bytes? Simply because we don't know of a case where more is needed. AVX alignment is 32 bytes; AVX-512 is 64. Fixes #8085, closes #8598.
| | * | | Merge pull request #9643 from rgommers/council-membersseberg2017-09-131-2/+10
| | |\ \ \ | | | |_|/ | | |/| | DOC: add new steering council members.
| | | * | DOC: add new steering council members.Ralf Gommers2017-09-091-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Discussion on the numpy-discussion mailing list on July 21-25 2017. [ci skip]
| * | | | BUG: Fix pinv on matrix stacksEric Wieser2017-09-121-0/+4
| |/ / / | | | | | | | | | | | | Fixes #8826
| * | | Merge pull request #6053 from ahaldane/multifield_structassignCharles Harris2017-09-091-0/+37
| |\ \ \ | | | | | | | | | | MAINT: struct assignment "by field position", multi-field indices return views
| | * | | MAINT: improve struct assignment, allow multi fields viewsAllan Haldane2017-09-071-0/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit attempts to make structure assignment more consistent, and then changes multi-field indices to return a view instead of a copy. Assignment between structures now works "by field position" rather than "by field name". Fixes #2353, fixes #6085, fixes #3351, fixes #6085, fixes #6314, fixes #2346, fixes #7058, fixes #3641, fixes #5994, fixes #7262, fixes #7493
| * | | | DOC: Simplify terminologyNathaniel J. Smith2017-09-081-16/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For some reason this file insisted on using multiple redundant terms for array dimensionality, and in particular liked the word "rank", which is confusing and very rarely used in my experience. This commit drops the word rank and removes a number of parenthetical "X (which is to say Y)" phrasings.
| * | | | DOC: clarify wording in tutorialNathaniel J. Smith2017-09-071-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In gh-9664, a user got confused by this wording, because they thought it was talking about the length of a vector as a distance in cartesian space versus the length of a vector as the number of elements it contains.
| * | | | Merge pull request #9658 from MSeifert04/fix_keyword_as_parameter_nameCharles Harris2017-09-061-0/+6
| |\ \ \ \ | | | | | | | | | | | | BUG: Fix usage of keyword "from" as argument name for "can_cast".
| | * | | | BUG: Fix usage of keyword "from" as argument name for "can_cast".Michael Seifert2017-09-061-0/+6
| | |/ / / | | | | | | | | | | | | | | | | | | | | Also removed inconsistency between the second argument name between documentation ("totype") and code ("to").
* | | | | ENH: Add the out parameter to stack tooEric Wieser2017-09-121-4/+4
| | | | |
* | | | | ENH: Add out argument to `concatenate`Eric Wieser2017-09-121-0/+4
|/ / / /
* | | | API: Make ``a.flat.__array__`` return a copy when ``a`` non-contiguous. (#9447)Charles Harris2017-09-051-0/+16
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * API: Make ``a.flat.__array__`` return copy if ``a`` non-contiguous. Previously an UPDATEIFCOPY array was returned when `a` was non-contiguous and writeable. Exposing that type of array does not work well with PyPy as there is no refcount, and hence the writeback never occurs. The copy in this case is set to non-writable to expose cases where a writeback was expected. At some future time the copy will be made writable. See gh-7054 for a discussion of this issue. It is not expected that this change will affect many people.
* | | DOC: release: add mention of msvc+gfortran to release notesPauli Virtanen2017-09-021-0/+15
|/ /
* | [DOC] add release note for f2py's dimension arraysNico Schlömer2017-08-281-0/+5
| |
* | Merge pull request #9613 from charris/update-1.14-notesCharles Harris2017-08-271-0/+5
|\ \ | | | | | | DOC: Update release notes for noncentral_f changes.
| * | DOC: Update release notes for noncentral_f changes.Charles Harris2017-08-271-0/+5
| | | | | | | | | | | | | | | | | | | | | The `dfnum` parameter is now required to be > 0 rather than the previous > 1. [ci skip]
View Lorry Controller Status