| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | MAINT: Directly grab `nan` from the input array | Bas van Beek | 2021-05-22 | 1 | -5/+7 |
| | | | | | Directly grab a nan-esque object from the input array, rather than constructing a new one from scratch | ||||
| * | BUG: Fixed an issue wherein `_nanmedian_small` would fail for ↵ | Bas van Beek | 2021-05-22 | 1 | -2/+4 |
| | | | | | `timedelta64`-based dtypes | ||||
| * | BUG: Fixed an issue wherein `nanmedian` could return an array with the wrong ↵ | Bas van Beek | 2021-05-22 | 1 | -1/+3 |
| | | | | | dtype | ||||
| * | DOC: Nupydoc format space before `:` in Parameters | Matthias Bussonnier | 2021-01-31 | 1 | -1/+1 |
| | | | | | Missing `s` in two spellings. | ||||
| * | Fix docstring cross-referencing | Albert Villanova del Moral | 2020-09-04 | 1 | -9/+9 |
| | | |||||
| * | MAINT: Remove unnecessary 'from __future__ import ...' statements | Jon Dufresne | 2020-01-03 | 1 | -2/+0 |
| | | | | | | As numpy is Python 3 only, these import statements are now unnecessary and don't alter runtime behavior. | ||||
| * | PERF: only copy input array in _replace_nan() if there are nans to replace | Christopher Whelan | 2019-12-02 | 1 | -1/+2 |
| | | |||||
| * | ENH: add OO->? loops, use np.compare(a, b, dtype=bool), add comments | mattip | 2019-10-30 | 1 | -1/+1 |
| | | |||||
| * | WIP, DEP, ENH: finish richcompare changes from 1.10 | mattip | 2019-10-29 | 1 | -1/+1 |
| | | |||||
| * | DOC: fix output argument reference | mattip | 2019-10-15 | 1 | -15/+15 |
| | | |||||
| * | DOC: Fixed dtype docs for var, nanvar. | Isaac Virshup | 2019-08-27 | 1 | -1/+1 |
| | | | | | | | | | | Docs incorrectly stated that the default output dtype was float32 when input array is of an integer dtype. However: ```python np.var(np.random.randint(0, 100, 1000)).dtype == np.dtype("f8") ``` | ||||
| * | Update nanfunctions.py | He Jia | 2019-06-14 | 1 | -0/+1 |
| | | |||||
| * | BUG: Increment stacklevel for warnings to account for NEP-18 overrides (#13589) | Stephan Hoyer | 2019-05-20 | 1 | -9/+16 |
| | | | | | | | | | | | | | | | * Increment stacklevel for warnings to account for NEP-18 overrides For NumPy functions that make use of `__array_function__`, the appropriate the stack level for warnings should generally be increased by 1 to account for the override function defined in numpy.core.overrides. Fixes GH-13329 * Update numpy/lib/type_check.py Co-Authored-By: Sebastian Berg <sebastian@sipsolutions.net> | ||||
| * | ENH: add _nan_mask function | Tyler Reddy | 2019-01-18 | 1 | -0/+27 |
| | | | | | | | | | | * add _nan_mask function and associated unit tests; this function is to be used to simplify reduction operations involving np.nan elements by using a mask with a ``where`` argument | ||||
| * | MAINT: address several reviewer comments | Tyler Reddy | 2018-12-14 | 1 | -2/+2 |
| | | |||||
| * | TST, DOC: enable refguide_check | Tyler Reddy | 2018-12-14 | 1 | -49/+52 |
| | | | | | | | | | * ported the refguide_check module from SciPy for usage in NumPy docstring execution/ verification; added the refguide_check run to Azure Mac OS CI * adjusted NumPy docstrings such that refguide_check passes | ||||
| * | MAINT: set preferred __module__ for numpy functions | Stephan Hoyer | 2018-10-23 | 1 | -1/+6 |
| | | |||||
| * | ENH: __array_function__ for np.lib, part 1 | Stephan Hoyer | 2018-10-08 | 1 | -0/+76 |
| | | | | | np.lib.arraypad through np.lib.nanfunctions | ||||
| * | DOC: reduce warnings and cleanup redundant c-api documentation | mattip | 2018-08-14 | 1 | -7/+9 |
| | | |||||
| * | Merge pull request #10199 from chunweiyuan/quantile | Stephan Hoyer | 2018-04-23 | 1 | -2/+108 |
| |\ | | | | | ENH: Quantile | ||||
| | * | ENH: Adding np.quantile() and np.nanquantile(). #10199 | Chun-Wei Yuan | 2018-04-16 | 1 | -2/+108 |
| | | | |||||
| * | | DOC: clear up warnings, fix matplotlib plot | mattip | 2018-04-17 | 1 | -7/+8 |
| |/ | |||||
| * | DOC: Fix type of axis in nanfunctions (#10421) | Tobias Fischer | 2018-02-03 | 1 | -14/+14 |
| | | |||||
| * | MAINT: Avoid repeated validation of percentiles in nanpercentile | Eric Wieser | 2017-12-18 | 1 | -18/+27 |
| | | | | | | | Previously, this would check the percentiles were 0 <= q <= 100 on every single slice of non-reduction axes. This also pulls the division by 100 to the top level, which makes things clearer, and also reduces the size of the stack trace | ||||
| * | DOC: give correct version of np.nansum change | Matthew Brett | 2017-12-19 | 1 | -1/+1 |
| | | | | | | | | nansum started returning 0 from all-nan slices in 1.9.0, according to my tests, not 1.8.1 or 1.8.2, which still return NaN. [skip ci] | ||||
| * | DOC: Fixup percentile docstring, from review in gh-9213 | Eric Wieser | 2017-12-10 | 1 | -19/+14 |
| | | | | | Updates the two docstrings to match. | ||||
| * | Make warnings for nanmin and nanmax consistent | Erik Quaeghebeur | 2017-10-23 | 1 | -1/+1 |
| | | |||||
| * | DOX: Fix mistaken word in nanprod docstring (#9703) | Robert T. McGibbon | 2017-09-17 | 1 | -1/+1 |
| | | |||||
| * | MAINT: Use moveaxis instead of rollaxis internally (#9475) | Nico Schlömer | 2017-08-10 | 1 | -1/+1 |
| | | | | | | Also add a hint to the documentation advising the use of moveaxis over rollaxis. Tests for rollaxis are left alone. | ||||
| * | Merge pull request #9220 from eric-wieser/simplify-nanfuncs | Charles Harris | 2017-06-22 | 1 | -43/+53 |
| |\ | | | | | MAINT: Factor out code duplicated by nanmedian and nanpercentile | ||||
| | * | MAINT: Factor out code duplicated by nanmedian and nanpercentile | Eric Wieser | 2017-06-21 | 1 | -43/+53 |
| | | | |||||
| * | | DOC: BLD: fix lots of Sphinx warnings/errors. | Ralf Gommers | 2017-06-10 | 1 | -1/+1 |
| |/ | |||||
| * | MAINT: Don't internally use the one-argument where | Eric Wieser | 2017-06-03 | 1 | -2/+2 |
| | | | | | nonzero is a clearer spelling | ||||
| * | MAINT: Combine similar branches | Eric Wieser | 2017-06-03 | 1 | -4/+1 |
| | | |||||
| * | BUG: Fix incorrect behavior of nanfunctions on object arrays | Eric Wieser | 2017-04-29 | 1 | -8/+16 |
| | | | | | Fixes gh-8974 and gh-9008 | ||||
| * | MAINT: Remove weird create-a-copy dance | Eric Wieser | 2017-04-27 | 1 | -6/+2 |
| | | |||||
| * | DOC: Fix typos in percentile (#8900) | Baurzhan Muftakhidinov | 2017-04-06 | 1 | -2/+2 |
| | | |||||
| * | ENH: retune apply_along_axis nanmedian cutoff | Julian Taylor | 2016-12-17 | 1 | -1/+2 |
| | | | | | | Old value was erroneously obtained on a sorted array which is a best case for the median of 3 pivoted introsort. | ||||
| * | Merge pull request #8364 from juliantaylor/masked-median-nan | Charles Harris | 2016-12-12 | 1 | -1/+1 |
| |\ | | | | | BUG: handle unmasked NaN in ma.median like normal median | ||||
| | * | ENH: update the small nanmedian threshold | Julian Taylor | 2016-12-12 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | The apply_along_axis path is significantly more expensive than currently accounted for in the check. Increase the minimum axis size from 400 to 1000 elements. Either apply_along_axis got more expensive over time or the original benchmarking was flawed. | ||||
| * | | BUG: fix nanpercentile not returning scalar with axis argument | Julian Taylor | 2016-12-11 | 1 | -1/+1 |
| |/ | | | | Closes gh-8220 | ||||
| * | DOC: change version references from x.y to x.y.z | Pierre de Buyl | 2016-09-07 | 1 | -1/+1 |
| | | |||||
| * | DOC: change Numpy to NumPy in dosctrings and comments | Pierre de Buyl | 2016-09-06 | 1 | -3/+3 |
| | | | | | The strings in error messages were left untouched | ||||
| * | ENH: Add stacklevel to all (or almost all) our function calls | Sebastian Berg | 2016-09-02 | 1 | -9/+9 |
| | | |||||
| * | ENH: Remove warning ignoring from nanfuncs | Sebastian Berg | 2016-09-02 | 1 | -39/+33 |
| | | | | | | | Comment mentions a speedup, but it seems unsure why it should be there. Instead use an error state in divide_by_count. Some extra complex warnings had to be ignored (but those seemed correct) | ||||
| * | ENH: adds np.nancumsum and np.nancumprod | Phillip J. Wolfram | 2016-03-24 | 1 | -15/+139 |
| | | | | | | | | | This PR adds an implementation of `nancumsum` and `nancumprod`. The actual function is a two-liner adapted from `nansum`. Its structure is adapted from PR: https://github.com/numpy/numpy/pull/5418/ | ||||
| * | DOC: Removed residual merge markup from previous commit | Mad Physicist | 2016-02-13 | 1 | -6/+0 |
| | | |||||
| * | BUG: many functions silently drop `keepdims` kwarg | Thomas A Caswell | 2016-02-07 | 1 | -39/+106 |
| | | | | | | | | | | | | | | | | | | | | | change test from `type(a) is not mu.ndarray` to `not isinstance(a, mu.ndarray)` Because every sub-class of ndarray is not guaranteed to implement `keepdims` as a kwarg, when wrapping these methods care must be taken. The previous behavior was to silently eat the kwarg when dealing with a sub-class of ndarray. Now, if `keepdims=np._NoValue` (the new default) it is not passed through to the underlying function call (so the default value of `keepdims` is now controlled by the sub-class). If `keepdims` is not `np._NoValue` then it is passed through and will raise an exception if the sub-class does not support the kwarg. A special case in nanvar was required to deal with `matrix` that previously relied on `fromnumeric` silently dropping `keepdims`. | ||||
| * | Merge pull request #7181 from madphysicist/doc-typos | Charles Harris | 2016-02-07 | 1 | -78/+81 |
| |\ | | | | | DOC: Updated minor typos in function_base.py and test_function_base.py | ||||
| | * | DOC: Updated minor typos in function_base.py and test_function_base.py | Joseph Fox-Rabinovitz | 2016-02-05 | 1 | -78/+81 |
| | | | |||||
 |
index : delta/python-packages/numpy.git | |
| github.com: numpy/numpy.git |
| summaryrefslogtreecommitdiff |