[FEA] Plumb edit_distance_matrix to Python

[beckernick]

We should plumb edit_distance_matrix to the Cython layer. This could potentially live in the cudf._lib.nvtext.edit_distance namespace alongside the canonical edit_distance.

For now, I'm not sure we if need to plumb this all the way up to the Python layer. See #4569 (comment) in which we decided similarly for find_multiple based on efficiency and utility. With list column support we could plumb it all the way up if desired.

It would be useful if we could analogously do for edit_distance_matrix what we can currently do for find_multiple.

from cudf._lib.strings import find_multiple
import cudf
import cupy as cp

a = cudf.Series(["rabbit","dog", "bird"])
b = cudf.Series(["a","b"])

out = find_multiple.find_multiple(a._column, b._column) # NumericalColumn output (contiguous buffer)
out = cp.asarray(out).reshape(-1, len(b)) # reshape to "actual" 2-d matrix representation
out
array([[ 1,  2],
       [-1, -1],
       [-1,  0]], dtype=int32)

[github-actions]

This issue has been marked rotten due to no recent activity in the past 90d. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed.

[beckernick]

This is still necessary, and likely should be plumbed up to the Python layer given list column support. Updating the title accordingly.

[davidwendt]

From libcudf, it looks like nvtext::edit_distance_matrix returns a lists column whereas cudf::strings::find_multiple returns a single array. It would not be difficult to change find_multiple return a lists column if that is helpful.

[github-actions]

This issue has been labeled inactive-90d due to no recent activity in the past 90 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed.