ajpotts

Create a shares_memory function to mirror numpy: https://numpy.org/doc/stable/reference/generated/numpy.shares_memory.html

- [ ] https://github.com/Bears-R-Us/arkouda/issues/3281 - [ ] https://github.com/Bears-R-Us/arkouda/issues/3287 - [ ] https://github.com/Bears-R-Us/arkouda/issues/3288 - [ ] https://github.com/Bears-R-Us/arkouda/issues/3289 - [ ] https://github.com/Bears-R-Us/arkouda/issues/3291 - [ ] https://github.com/Bears-R-Us/arkouda/issues/3295 - [ ] https://github.com/Bears-R-Us/arkouda/issues/3296 - [...

binops to handle np.nan correctly

1

This is what I am seeing" ``` python Out[4]: array([False True True True True]) In [5]: x = ak.array([np.nan, 1.0]) In [6]: x == x Out[6]: array([False True]) In [7]:...

Extend to accept multiple aggregations as a dictionary

1

e.g. {col1: agg1, col2: [agg2a, agg2b], ...}. See pandas usage. Only valid with df.groupby() result.

Investigate discrepancies between pandas and arkouda MultiIndex APIs: https://pandas.pydata.org/docs/reference/api/pandas.MultiIndex.html In particular, pandas has `level` and `codes` attributes that are similar to arkouda Groupby `categories` and `codes` attributes. ``` python In...

Add skipna option to DataFrame.groupby.first to mirror pandas: https://pandas.pydata.org/docs/reference/api/pandas.core.groupby.DataFrameGroupBy.first.html

Implement an `array_equal` function to mirror numpy: https://numpy.org/doc/stable/reference/generated/numpy.array_equal.html

Verify that all tests have been moved to the PROTO_tests directory and then deprecate the tests.

I could not find any unit tests for `argsort` or `coargsort`.

bug in array_api.matmul for arrays of different shapes

2

**Describe the bug** Matrix multiplication does not work for all shapes where it should work. I suspect it's b/c the `broadcast_dims` function is written to support bin ops which have...