Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / ml31415/numpy-groupies issues and pull requests

#89 - Reduce skipped tests

Pull Request - State: closed - Opened by ml31415 2 months ago

#88 - Less skipped tests

Pull Request - State: closed - Opened by ml31415 2 months ago

#87 - Improve publishing to pypi

Pull Request - State: closed - Opened by ml31415 2 months ago

#86 - Routine maintenance

Pull Request - State: closed - Opened by ml31415 2 months ago

#85 - Exploring GPU acceleration using CuPy for enhanced performance

Issue - State: closed - Opened by zjzjwang 2 months ago - 2 comments
Labels: question

#84 - Any recommended way to broadcast result back to the shape of "a"?

Issue - State: closed - Opened by ancri 7 months ago - 3 comments
Labels: question

#83 - Adding py3.12

Pull Request - State: closed - Opened by ml31415 9 months ago - 2 comments

#82 - Different package file names

Issue - State: open - Opened by ml31415 9 months ago

#81 - Cleanup

Pull Request - State: closed - Opened by ml31415 9 months ago

#80 - Add numpy implementation of cumprod

Pull Request - State: closed - Opened by Illviljan 12 months ago - 3 comments

#79 - nancumsum result for numba and numpy does not match

Issue - State: closed - Opened by Illviljan 12 months ago - 1 comment
Labels: bug

#78 - Some optimizations for numpy aggregations

Pull Request - State: closed - Opened by dcherian about 1 year ago

#77 - Use faster group_idx creation when axis == -1

Pull Request - State: closed - Opened by dcherian over 1 year ago - 2 comments

#76 - Move to micromamba

Pull Request - State: closed - Opened by dcherian over 1 year ago - 1 comment

#75 - CI conda build fails

Issue - State: closed - Opened by ml31415 over 1 year ago - 1 comment

#74 - only 1d indexing currentlysupported with axis arg

Issue - State: open - Opened by joshua-gould over 1 year ago
Labels: bug

#73 - rerun benchmarks with numpy 1.25

Issue - State: closed - Opened by dcherian over 1 year ago - 4 comments

#72 - Fix pytest failure without numba.

Pull Request - State: closed - Opened by sebastic over 1 year ago - 2 comments

#71 - Fix versioneer configuration.

Pull Request - State: closed - Opened by sebastic over 1 year ago - 1 comment

#70 - Repo name change?

Issue - State: closed - Opened by cmdupuis3 over 1 year ago - 2 comments

#69 - AggregateGeneric fails with np.median

Issue - State: open - Opened by ml31415 over 1 year ago
Labels: bug

#68 - Add pypi release workflow

Issue - State: closed - Opened by dcherian over 1 year ago - 6 comments

#67 - Initial draft and benchmarking experiments with polars

Pull Request - State: closed - Opened by ml31415 over 1 year ago

#66 - release

Issue - State: closed - Opened by dcherian over 1 year ago - 3 comments

#65 - Multiple threads usage in aggregate_np

Issue - State: closed - Opened by makihor over 1 year ago - 3 comments

#64 - Generic non reduce function on aggregate

Issue - State: closed - Opened by josephnowak almost 2 years ago - 1 comment

#63 - Add CuPy support

Pull Request - State: open - Opened by dcherian almost 2 years ago - 1 comment

#62 - Remove numba from requirements

Pull Request - State: closed - Opened by dcherian almost 2 years ago

#61 - Switch back to ravel_multi_index.

Pull Request - State: closed - Opened by dcherian about 2 years ago

#60 - Different results in V0.9.14 than in V0.9.16

Issue - State: closed - Opened by josephnowak about 2 years ago - 1 comment

#59 - Fix custom callable reductions.

Pull Request - State: closed - Opened by dcherian about 2 years ago

#58 - Update CI

Pull Request - State: closed - Opened by dcherian about 2 years ago

#57 - Another argreduction fix.

Pull Request - State: closed - Opened by dcherian about 2 years ago - 1 comment

#56 - new release

Issue - State: closed - Opened by dcherian about 2 years ago

#55 - Trapz implementation for numba

Pull Request - State: closed - Opened by bmorris3 over 2 years ago - 3 comments

#54 - trapz support?

Issue - State: open - Opened by bmorris3 over 2 years ago - 1 comment
Labels: enhancement

#53 - Make var/std work with NaN fill_value

Pull Request - State: closed - Opened by dcherian over 2 years ago

#52 - isort

Pull Request - State: closed - Opened by dcherian over 2 years ago

#51 - Switch from ravel_multi_index to a faster vectorized alternative.

Pull Request - State: closed - Opened by dcherian over 2 years ago - 2 comments

#50 - Don't convert duck arrays to numpy arrays

Pull Request - State: closed - Opened by dcherian over 2 years ago

#49 - autopep8

Pull Request - State: closed - Opened by dcherian over 2 years ago - 1 comment

#48 - Test for reduction along axis + fix arg*, cum* for form3

Pull Request - State: closed - Opened by dcherian over 2 years ago

#47 - Add Github Actions CI

Pull Request - State: closed - Opened by dcherian over 2 years ago - 1 comment

#46 - Faster input validation

Pull Request - State: closed - Opened by dcherian over 2 years ago

#45 - "mean" promotes float32 to float64

Issue - State: closed - Opened by dcherian over 2 years ago - 3 comments

#44 - bug with nanvar, nD arrays with all-NaN rows, axis=-1

Issue - State: closed - Opened by dcherian over 2 years ago

#43 - Support median

Issue - State: open - Opened by dcherian over 2 years ago
Labels: enhancement

#42 - aggregate downcasts duck arrays to ndarray

Issue - State: closed - Opened by Illviljan over 2 years ago - 1 comment

#41 - bug with argmax and nd array

Issue - State: closed - Opened by dcherian almost 3 years ago

#40 - bug with max and -np.inf

Issue - State: closed - Opened by dcherian almost 3 years ago - 2 comments

#39 - nanarg* broken

Issue - State: closed - Opened by dcherian almost 3 years ago - 2 comments

#38 - debugged setting size automatically when ndim_idx > 1

Pull Request - State: closed - Opened by yulkang almost 3 years ago - 2 comments

#37 - (Potentially) faster-than-Pandas pd.factorize, demo code for discussion

Issue - State: closed - Opened by ianozsvald about 3 years ago - 1 comment

#36 - question: method for getting argtopk

Issue - State: closed - Opened by welaunchcn about 3 years ago - 1 comment

#35 - aggregate_nb should make use of all available CPU cores

Issue - State: closed - Opened by lanougue about 3 years ago - 7 comments

#34 - npg.aggregate(group_idx, 1) fails with multidimensional group

Issue - State: closed - Opened by davidhbrann over 3 years ago - 2 comments

#33 - Set the length of the return value

Issue - State: closed - Opened by acycliq over 3 years ago

#32 - Unreasonable default fill_values

Issue - State: open - Opened by max-sixty over 3 years ago - 7 comments

#31 - Multi column agg

Issue - State: closed - Opened by ghost over 3 years ago - 4 comments

#30 - [openSUSE]: Python 2.7.17, E NameError: name 'aggregate_numba' is not defined

Issue - State: closed - Opened by kevinsmia1939 almost 4 years ago - 2 comments

#29 - Grouping by unsigned integers throws an error

Issue - State: closed - Opened by ivirshup about 4 years ago - 1 comment

#28 - Fix versionning from source archive

Issue - State: closed - Opened by ccoulombe about 4 years ago - 6 comments

#27 - BSD license should probalby be in included in manifest

Issue - State: closed - Opened by detrout over 4 years ago - 2 comments

#26 - Is there a way to fill the array with an array result?

Issue - State: closed - Opened by math-artist over 4 years ago - 1 comment

#25 - Multiple Input

Issue - State: closed - Opened by firmai over 4 years ago - 1 comment

#24 - fastfunc comparison, prescribe output format?

Issue - State: closed - Opened by nschloe almost 5 years ago - 3 comments

#23 - Equivalence between accumarray and aggregate?

Issue - State: closed - Opened by cdoucet about 5 years ago - 9 comments

#22 - func='mean', fill_value=nan returns all nans

Issue - State: closed - Opened by adbull over 5 years ago - 1 comment

#21 - Direct (and optimized?) support for labeled list of lists

Issue - State: closed - Opened by e-champenois over 5 years ago - 1 comment
Labels: enhancement

#20 - sequence too large

Issue - State: closed - Opened by brealisty over 5 years ago

#19 - Consider adding an optional output of the unique indices?

Issue - State: closed - Opened by yuxiangw almost 6 years ago - 3 comments

#17 - allow float and negatives for group_idx

Issue - State: closed - Opened by cardosan over 7 years ago - 2 comments

#16 - Python 3 support

Issue - State: closed - Opened by slinnarsson over 7 years ago - 2 comments

#15 - aggregate numba - optimise cache usage

Issue - State: closed - Opened by d1manson over 7 years ago - 2 comments
Labels: enhancement

#14 - invalid test_requires in setup.py

Issue - State: closed - Opened by anntzer almost 8 years ago - 1 comment

#13 - Please add a license

Issue - State: closed - Opened by shoyer about 8 years ago - 11 comments

#12 - len returns dtype object instead of int

Issue - State: closed - Opened by ml31415 over 8 years ago - 3 comments
Labels: bug

#11 - aggregate - check performance of scipy ndimage

Issue - State: closed - Opened by d1manson almost 9 years ago - 1 comment
Labels: question

#10 - setup.py file

Pull Request - State: closed - Opened by eendebakpt almost 9 years ago - 6 comments

#9 - Improve complex support in all aggregate implementations

Issue - State: open - Opened by d1manson almost 9 years ago
Labels: enhancement

#8 - publicise the package

Issue - State: closed - Opened by d1manson about 9 years ago - 17 comments

#7 - Optimizing C implementation for fortran arrays

Issue - State: closed - Opened by ml31415 about 9 years ago - 1 comment
Labels: enhancement

#6 - Fix numba implementation of aggregate

Issue - State: closed - Opened by ml31415 about 9 years ago - 18 comments
Labels: bug

#5 - Stabilize API for multi_cumsum and multi_range

Issue - State: closed - Opened by ml31415 about 9 years ago - 6 comments
Labels: enhancement

#4 - grouped_sort

Issue - State: closed - Opened by ml31415 about 9 years ago - 8 comments
Labels: enhancement

#3 - specify multiple aggregation functions

Issue - State: closed - Opened by d1manson about 9 years ago - 3 comments
Labels: enhancement

#2 - Create a single API/test/benchmarking but with multiple independent implementations

Issue - State: closed - Opened by d1manson about 9 years ago - 61 comments

#1 - what about numpy's ufunc.at and bincounts?

Issue - State: closed - Opened by d1manson over 9 years ago - 22 comments