summaryrefslogtreecommitdiff
path: root/numpy
Commit message (Collapse)AuthorAgeFilesLines
...
| * | | MAINT: types.TypeType does not ever need to be usedEric Wieser2017-08-051-12/+6
| |/ /
* | | MAINT: Remove `level=` keyword from test arguments.Charles Harris2017-08-058-173/+164
|/ / | | | | | | | | | | I don't know what that argument was used for, but it showis up in old tests and is not explicitly used within the tests. I assume it was part of an old testing framework and is now longer needed.
* | MAINT/DOC: Use builtin when np.{x} is builtins.{x}.Eric Wieser2017-08-0539-326/+326
| | | | | | | | | | | | | | This is the case for x in {int, bool, str, float, complex, object}. Using the np.{x} version is deceptive as it suggests that there is a difference. This change doesn't affect any external behaviour. The `long` type is missing in python 3, so np.long is still useful
* | Merge pull request #9469 from charris/fix-true-divideCharles Harris2017-08-046-43/+143
|\ \ | | | | | | BUG: Fix true_divide when dtype=np.float64 specified.
| * | BUG: Fix true_divide when dtype=np.float64 specified.Charles Harris2017-08-036-43/+143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The problem here was due to the signature matching when dtype is specified. A signature matches when 1) The output dtype matches the specified dtype 2) The input arguments can be cast to the signatures dtypes using casting="same_kind" The signature 'bb->d' was present for true_divide, consequently when dtype=np.float64 was specified, all integer inputs were cast to int8. The solution adopted here was to remove all signatures/loops for true_divide that had integer inputs. Lesser measures broke a lot of existing code that relied on the behavior for other functions Closes #3484. Again.
* | | Merge pull request #9457 from eric-wieser/ma-ufunc-fixesCharles Harris2017-08-041-27/+21
|\ \ \ | |/ / |/| | MAINT: Add a common subclass to all the masked ufunc wrappers
| * | MAINT: Add a common subclass to all the masked ufunc wrappersEric Wieser2017-07-241-27/+21
| | |
* | | Merge pull request #9492 from charris/add-chebyshev-interpolationCharles Harris2017-08-013-2/+163
|\ \ \ | | | | | | | | ENH: Add np.polynomial.chebyshev.chebinterpolate function.
| * | | MAINT: Rename chebinterp to chebinterpolation and add test.Charles Harris2017-07-293-43/+126
| | | | | | | | | | | | | | | | | | | | | | | | | | | | * Rename chebinterp to chebinterpolation as suggested. * Make some fixes to the Chebyshev class function. * Refactor TestInterpolation. * Add test for the Chebyshev.interpolation class function.
| * | | ENH: Add np.polynomial.chebyshev.chebinterp functionLukas Mericle2017-07-242-1/+79
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The chebinterp function interpolates a function at the Chebyshev points of the first kind using a Chebyshev series. The resulting interpolation approximates a polynomial fit minimizing the L_inf norm. This function is useful e.g. for estimating roots before running a root-finding function.
* | | | Merge pull request #9482 from madphysicist/iterative-diffEric Wieser2017-08-012-23/+84
|\ \ \ \ | | | | | | | | | | MAINT: Make diff iterative instead of recursive
| * | | | MAINT: Changed diff to use iterative instead of recursive approachJoseph Fox-Rabinovitz2017-07-312-23/+84
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TST: Added tests for `n` parameter Added test for `datetime64` type change Added tests for axis normalization Added test for subtype handling DOC: Minor updates to docs: Added explanation for `n==0` Added documentation describing `datetime64` handling Updated formatting Added call to normalize_axis_index
* | | | | MAINT: Move variables into deepest relevant scope, for clarityEric Wieser2017-07-311-5/+6
| | | | |
* | | | | Merge pull request #9499 from eric-wieser/fix-view-decrefsCharles Harris2017-07-311-51/+64
|\ \ \ \ \ | |/ / / / |/| | | | MAINT/BUG: Improve error messages for dtype reassigment, fix missing DECREFs
| * | | | MAINT/BUG: Improve error messages for dtype reassigment, fix missing DECREFEric Wieser2017-07-311-51/+64
| | | | |
* | | | | DOC: fix versionadded in docstring for moveaxisStephan Hoyer2017-07-301-1/+1
| | | | |
* | | | | TST: Add filter for new Py3K warning in python 2Sebastian Berg2017-07-291-2/+6
|/ / / / | | | | | | | | | | | | | | | | Python 2.7.13 added a new py3 compatibility DeprecationWarning we did not catch at the moment.
* | | | Merge pull request #9470 from wufangjie/setxor1dJaime2017-07-281-4/+1
|\ \ \ \ | | | | | | | | | | MAINT: Make `setxor1d' a bit clearer and speed it up
| * | | | MAINT: Make `setxor1d' a bit clearer and speed it upwufangjie2017-07-281-5/+1
| | | | |
| * | | | make `setxor1d' a bit clear and speed upwufangjie2017-07-271-2/+3
| | |_|/ | |/| | | | | | We need to find the index which is not the same with the left and right, I think np.logical_and's meaning is more clear and I test this got a speed up
* | | | BLD: remove -xhost flag from IntelFCompiler. Closes gh-9042.Ralf Gommers2017-07-271-1/+1
|/ / / | | | | | | | | | | | | Note that this was discussed extensively in gh-7287, but removing this -xhost flag was missed in the PR that closed that issue.
* | | Merge pull request #9454 from eric-wieser/remove-pointless-codeCharles Harris2017-07-241-12/+10
|\ \ \ | |_|/ |/| | MAINT: Remove branch in __array__ where if and else were the same
| * | MAINT: Remove branch in __array__ where if and else were the sameEric Wieser2017-07-241-12/+10
| | |
* | | Merge pull request #9452 from eric-wieser/new-style-classesCharles Harris2017-07-246-17/+17
|\ \ \ | |_|/ |/| | MAINT: Use new-style classes on 2.7
| * | MAINT: Use new-style classes on 2.7Eric Wieser2017-07-246-17/+17
| |/ | | | | | | Deliberately avoids tests, to prevent introducing a failure on old-style classes later.
* | Merge pull request #9434 from charris/remove-unittest-dependenciesCharles Harris2017-07-2481-1903/+1909
|\ \ | |/ |/| Remove unittest dependencies
| * TST: Remove unittest dependencies in numpy/f2py/tests.Charles Harris2017-07-242-9/+7
| |
| * TST: Remove unittest dependencies in numpy/tests.Charles Harris2017-07-241-23/+23
| |
| * TST: Remove unittest dependencies in numpy/random/tests.Charles Harris2017-07-242-30/+32
| |
| * TST: Remove unittest dependencies in numpy/polynomial/tests.Charles Harris2017-07-248-78/+85
| |
| * TST: Remove unittest dependencies in numpy/ma/tests.Charles Harris2017-07-247-594/+583
| |
| * TST: Remove unittest dependencies in numpy/linalg/tests.Charles Harris2017-07-242-28/+28
| |
| * TST: Remove unittest dependencies in numpy/lib/tests.Charles Harris2017-07-2417-328/+327
| |
| * TST: Remove unittest dependencies in numpy/fft/tests.Charles Harris2017-07-242-22/+26
| |
| * TST: Remove unittest dependencies in numpy/core/tests.Charles Harris2017-07-2430-685/+689
| | | | | | | | | | Also properly naming tests to begin with test, and using camelcase for class names.
| * TST: Remove unittest dependencies in numpy/distutils/testsCharles Harris2017-07-176-82/+84
| | | | | | | | Also rename some classes that should not begin 'Test'.
| * TST: Remove unittest dependencies in matrixlib/tests.Charles Harris2017-07-174-24/+25
| |
* | Merge pull request #9446 from charris/fix-gcc7.1-compileseberg2017-07-211-3/+3
|\ \ | | | | | | BUG: Inlined functions must be defined somewhere.
| * | BUG: Inlined functions must be defined somewhere.Charles Harris2017-07-211-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | Some inline functions in numpy/core/src/multiarray/mapping.c were defined as `NPY_NO_EXPORT NPY_INLINE` with the result that gcc 7.1 was looking for an external definition. Possibly some of those functions were no longer being inlined. The fix was to replace `NPY_NO_EXPORT` with `static`. A grep of the numpy source files turned up no further uses of that unfortunate combination.
* | | Merge pull request #9400 from eric-wieser/ufunc-namesCharles Harris2017-07-203-42/+30
|\ \ \ | | | | | | | | MAINT: Further unify handling of unnamed ufuncs
| * | | MAINT: Further unify handling of unnamed ufuncsEric Wieser2017-07-113-42/+30
| | | | | | | | | | | | | | | | Follows on from gh-8876
* | | | Merge pull request #9411 from eric-wieser/gradient-dtypesCharles Harris2017-07-202-27/+27
|\ \ \ \ | | | | | | | | | | MAINT/BUG: improve gradient dtype handling
| * | | | MAINT: Use dtypes, not typecodes, and remove special casing for timedeltaEric Wieser2017-07-131-27/+21
| | | | |
| * | | | BUG: float16 is promoted to float64 by gradientEric Wieser2017-07-132-1/+7
| | | | | | | | | | | | | | | | | | | | This isn't the case for `diff`
* | | | | Merge pull request #9412 from mattip/fix-9404Eric Wieser2017-07-203-8/+31
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | BUG: Check for exception in sort functions Fixes gh-3879 Fixes gh-9404
| * | | | | BUG: Check for exception in sort functions, add testsmattip2017-07-173-8/+31
| |/ / / /
* | | | | Merge pull request #8920 from juliantaylor/more-cache-allocJulian Taylor2017-07-2015-88/+110
|\ \ \ \ \ | |_|_|/ / |/| | | | ENH: use caching memory allocator in more places
| * | | | MAINT: document minimum dim cache size == 2Julian Taylor2017-07-111-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | | Remove unlikely, as it is being used now more it may not be correct anymore.
| * | | | ENH: use caching memory allocator in more placesJulian Taylor2017-07-1114-84/+103
| | |_|/ | |/| | | | | | | | | | | | | | In particular use it in PyArray_IntpConverter which is used in many function entrypoints.
* | | | MAINT: Make the refactor suggested in prepare_index (#8278)Eric Wieser2017-07-181-89/+201
| | | | | | | | | | | | | | | | | | | | | | | | | | | | MAINT: Make the refactor suggested in prepare_index This refactors the tuple creation part of the index creation (and actually removes the usage of a tuple itself in favor of a raw data array). Making the index preparation function shorter and more specific to the important task.
View Lorry Controller Status