DOC: argmin, argmax, argort, etc. should link to np.take_along_axis in their documentation
#6078
Comments
|
Indeed, this problem has also flummoxed me. Here's the helper function I wrote, which may be slightly more interpretable: def argsort_indices(a, axis=-1):
"""Like argsort, but returns an index suitable for sorting the
the original array even if that array is multidimensional
"""
a = np.asarray(a)
ind = list(np.ix_(*[np.arange(d) for d in a.shape]))
ind[axis] = a.argsort(axis)
return tuple(ind) |
|
@shoyer - yours is definitely easier to read, in that it leaves the proper shaping to |
I guess it's less general in that it only works for |
|
I think it would be lot better than today's implementation if the np.unravel_index(np.argsort(a, axis=None), a.shape) |
As a rule, NumPy doesn't change existing behavior. So this will need to be new API of some sort. |
|
@shoyer Oh that's true. |
|
I think this issue is now irrelevant, since we have |
np.take_along_axis in their documentation
In #6075, I wondered whether it might be possible to
and @jaimefrio suggested that it might be simpler to have
That indeed seems better, so I'm raising this as a separate issue. The code linked to above could be used as a template on how to generate the index tuple.
The text was updated successfully, but these errors were encountered: