| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | BUG: Fix in1d for empty integer array as input (#21842) | Miles Cranmer | 2022-06-24 | 1 | -0/+7 |
| | | | | | | | | | | | | * TST: Tests empty input to in1d of various types * MAINT: Skip table method on empty input array * MAINT: Check array size before call to min * MAINT: Return early for kind="table" and empty ar2 * TST: Test other dtypes of empty arrays | ||||
| * | MAINT: Fix edgecase for bool containers | MilesCranmer | 2022-06-22 | 1 | -2/+2 |
| | | |||||
| * | DOC: Rephrase docstring of in1d/isin | MilesCranmer | 2022-06-22 | 1 | -2/+2 |
| | | |||||
| * | MAINT: Remove positionality of kind in isin | MilesCranmer | 2022-06-22 | 1 | -2/+2 |
| | | |||||
| * | MAINT: Minor suggestions from code review | Miles Cranmer | 2022-06-22 | 1 | -4/+3 |
| | | | | Co-authored-by: Sebastian Berg <sebastian@sipsolutions.net> | ||||
| * | DOC: Notes on `kind` to in1d/isin docstring | MilesCranmer | 2022-06-20 | 1 | -10/+18 |
| | | |||||
| * | DOC: `assume_unique` does not affect table method | MilesCranmer | 2022-06-20 | 1 | -4/+8 |
| | | |||||
| * | DOC: Improve clarity of in1d docstring | MilesCranmer | 2022-06-20 | 1 | -6/+6 |
| | | |||||
| * | MAINT: change kind names for in1d | MilesCranmer | 2022-06-18 | 1 | -23/+23 |
| | | | | | - Switch dictionary->table, mergesort->sort | ||||
| * | DOC: Fix list format for sphinx | Miles Cranmer | 2022-06-18 | 1 | -9/+11 |
| | | |||||
| * | DOC: Add missing indent in docstring | Miles Cranmer | 2022-06-18 | 1 | -1/+1 |
| | | |||||
| * | DOC: Fix formatting issues in docstring | MilesCranmer | 2022-06-17 | 1 | -4/+2 |
| | | |||||
| * | MAINT: RuntimeError for unexpected integer overflow | MilesCranmer | 2022-06-17 | 1 | -0/+7 |
| | | |||||
| * | MAINT: Clean up integer overflow check in in1d | MilesCranmer | 2022-06-17 | 1 | -11/+5 |
| | | |||||
| * | MAINT: Clean up memory checking for in1d | MilesCranmer | 2022-06-17 | 1 | -11/+9 |
| | | |||||
| * | MAINT: Protect against integer overflow in in1d | MilesCranmer | 2022-06-17 | 1 | -3/+6 |
| | | |||||
| * | MAINT: kind now uses "mergesort" instead of "sort" | MilesCranmer | 2022-06-17 | 1 | -9/+9 |
| | | |||||
| * | MAINT: Switch parameter name to 'kind' over 'method' | MilesCranmer | 2022-06-17 | 1 | -24/+28 |
| | | |||||
| * | MAINT: Switch to old in1d for large memory usage | MilesCranmer | 2022-06-17 | 1 | -13/+18 |
| | | |||||
| * | DOC: Describe memory considerations in in1d/isin | MilesCranmer | 2022-06-10 | 1 | -10/+22 |
| | | |||||
| * | DOC: Clean up isin docstring | MilesCranmer | 2022-06-10 | 1 | -3/+3 |
| | | |||||
| * | MAINT: bool instead of np.bool_ dtype | MilesCranmer | 2022-06-10 | 1 | -6/+6 |
| | | |||||
| * | DOC: Improve docstring explanation | MilesCranmer | 2022-06-10 | 1 | -2/+2 |
| | | |||||
| * | MAINT: Formatting changes for in1d | Miles Cranmer | 2022-06-10 | 1 | -2/+2 |
| | | | | Co-authored-by: Hameer Abbasi <2190658+hameerabbasi@users.noreply.github.com> | ||||
| * | MAINT: Fix error message | MilesCranmer | 2022-06-10 | 1 | -1/+2 |
| | | |||||
| * | MAINT: Fix use of dispatcher for isin | MilesCranmer | 2022-06-10 | 1 | -2/+2 |
| | | |||||
| * | DOC: Describe default in docstring | MilesCranmer | 2022-06-10 | 1 | -2/+2 |
| | | |||||
| * | DOC: Specify constraints of method in docstring | MilesCranmer | 2022-06-10 | 1 | -2/+4 |
| | | |||||
| * | Add check for methods | MilesCranmer | 2022-06-10 | 1 | -0/+10 |
| | | |||||
| * | MAINT: Change `_slow_integer` parameter to `method` | MilesCranmer | 2022-06-10 | 1 | -16/+31 |
| | | |||||
| * | MAINT: Fix linting errors in in1d tests | MilesCranmer | 2022-06-09 | 1 | -1/+2 |
| | | |||||
| * | TST: Extend np.in1d tests to old algorithm | MilesCranmer | 2022-06-09 | 1 | -7/+19 |
| | | | | | | - Add flag ``_slow_integer`` to np.isin/np.in1d to force the use of the old isin/in1d algorithm for integers. | ||||
| * | MAINT: Check for overflow in integral np.isin | MilesCranmer | 2022-06-09 | 1 | -9/+16 |
| | | | | | | | | | | | | - Before this fix, if, e.g., int8(-100) and int8(100) are the min and max, respectively, of ar2, ar2_range would experience an integer overflow and optimal_parameters would be an invalid value. ``if optimal_parameters:`` would skip the fast integer algorithm regardless, but warnings would be generated when they are unnecessary. - np.log10(... + 1) was changed to np.log10(... + 1.0) to avoid further potential problems from overflow here. - See https://github.com/numpy/numpy/pull/12065 for discussion. | ||||
| * | MAINT: Optimize np.isin for boolean arrays | MilesCranmer | 2022-06-09 | 1 | -0/+6 |
| | | | | | | - This change converts boolean input to numpy.isin and numpy.in1d into uint8, allowing the function to use the new optimized code | ||||
| * | MAINT: Optimize np.isin for integer arrays | MilesCranmer | 2022-06-09 | 1 | -0/+40 |
| | | | | | | | - This optimization indexes with an intermediary boolean array to speed up numpy.isin and numpy.in1d for integer arrays over a range of optimal parameters which are calculated. | ||||
| * | MAINT: Fixup `unique`s `equal_nan` kwarg to match `np.array_equal` | Sebastian Berg | 2022-06-01 | 1 | -11/+11 |
| | | | | | | | | | The other similar arguments in the testing functions and `np.array_equal` use `equal_nan` and not `equal_nans` (with plural s). So it seems we should use that here as well for consistency. Does some very minor style fixes otherwise. | ||||
| * | MAINT: Update multiline indentations | Jason Thai | 2022-06-01 | 1 | -1/+1 |
| | | | | Co-authored-by: Matti Picus <matti.picus@gmail.com> | ||||
| * | DOC: Update 21623 release documentation | Jason Thai | 2022-05-29 | 1 | -1/+1 |
| | | | | Co-authored-by: Matti Picus <matti.picus@gmail.com> | ||||
| * | DOC: Added releasenote for 21623 | Jason Thai | 2022-05-28 | 1 | -4/+4 |
| | | |||||
| * | ENH: Added equal-nans kwarg to np.unique | Jason Thai | 2022-05-28 | 1 | -7/+11 |
| | | |||||
| * | DOC: misc fixes: | Matthias Bussonnier | 2022-03-21 | 1 | -1/+1 |
| | | | | | | | | | | | - improper urls - a few use of single bakcticks where many are expected - Proper numpydoc format [ci skip] [skip azp] [skip circle] | ||||
| * | BUG: Do not raise deprecation warning for all nans in unique (#19301) | Thomas J. Fan | 2021-06-29 | 1 | -1/+3 |
| | | | | | | This PR adjusts np.unique for the edge cases where all values are nan. Fixes gh-19300 | ||||
| * | BUG: Fix unique handling of nan entries. (#18070) | Filip Trojan | 2021-02-12 | 1 | -1/+20 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * benchmark bench_lib.Unique added * extended test_unique_1d * modify _unique1d * extend test with return_index, return_inverse and return_counts parameters * documentation updated * Update numpy/lib/arraysetops.py Co-authored-by: Bas van Beek <43369155+BvB93@users.noreply.github.com> * full coverage of nan types Co-authored-by: Bas van Beek <43369155+BvB93@users.noreply.github.com> * added tests for the datetime like dtypes * nan as vector of length 1 * use aux[-1] as nan, ..versionchanged, release note * for complex arrays all NaN values are considered equivalent Co-authored-by: filip_trojan <Tarantula2018> Co-authored-by: Bas van Beek <43369155+BvB93@users.noreply.github.com> | ||||
| * | Implement different approach to fix bug | AngelGris | 2021-02-08 | 1 | -20/+8 |
| | | |||||
| * | BUG: np.in1d bug on the object array (issue 17923) | AngelGris | 2021-02-08 | 1 | -4/+20 |
| | | |||||
| * | BUG: Fixes incorrect error message in numpy.ediff1d (#17457) | stuartarchibald | 2020-10-05 | 1 | -1/+1 |
| | | | | | | | Fixes the error message reported by ediff1d in the case of an invalid to_begin and updates tests to assert validity. Co-authored-by: Ross Barnowski <rossbar@berkeley.edu> | ||||
| * | DOC: Add arraysetops to autosummary. | Ross Barnowski | 2020-09-29 | 1 | -16/+5 |
| | | | | | | | * Generate stub for module, fixes 7 broken links * Modify arraysetops module docstring w/ rst formatting * Remove listing of functions from module docstring. | ||||
| * | MAINT: Chain some exceptions in arraysetops. (#17132) | Raghav Khanna | 2020-08-23 | 1 | -3/+3 |
| | | | | | | | | | | | | * MAINT: chain type error line 303 arraysetops.py * MAINT: chain axis error line 281 arraysetops.py * Revert "MAINT: chain axis error line 281 arraysetops.py" This reverts commit f6e1f544bebb7e5346cb83bdac9756be6ee9f4f6. * MAINT: chain axis error line 281 arraysetops.py to be from None | ||||
| * | DOC: Improve intersect1d docstring (#16420) | Dima Kogan | 2020-08-05 | 1 | -1/+3 |
| | | | | | | | The docstring now says what to expect if you call intersect1d(assume_unique=True) but pass in non-unique data. Co-authored-by: Ross Barnowski <rossbar@berkeley.edu> | ||||
| * | DOC: add example to unique() and make connection to repeat() (#16743) | Allen Downey | 2020-07-07 | 1 | -1/+13 |
| | | | | I suggest adding `repeat` to the "See also" section of `unique` (and vice versa) since each is the inverse of the other. | ||||
 |
index : delta/python-packages/numpy.git | |
| github.com: numpy/numpy.git |
| summaryrefslogtreecommitdiff |