DOC: Mention that expand_dims and squeeze are inverses #8737
Conversation
numpy/core/fromnumeric.py
Outdated
| @@ -1162,6 +1162,10 @@ def squeeze(a, axis=None): | |||
| dimensions of length 1 removed. This is always `a` itself | |||
| or a view into `a`. | |||
|
|
|||
| See Also | |||
| -------- | |||
| expand_dims : The inverse operation, adding singleton dimensions | |||
There was a problem hiding this comment.
It's not quite the inverse, as it only inserts one axis at a time.
| @@ -255,6 +255,7 @@ def expand_dims(a, axis): | |||
|
|
|||
| See Also | |||
| -------- | |||
| squeeze : The inverse operation, removing singleton dimensions | |||
There was a problem hiding this comment.
Maybe, "removes all singleton dimensions".
There was a problem hiding this comment.
What I mean here is that np.squeeze(arr, axis=a) and np.expand_dims(arr, axis=a) are inverses. I guess this only applies to one of the two overloads
There was a problem hiding this comment.
Might also mention reshape in both contexts as it allows adding and removing axis.
There was a problem hiding this comment.
I personally like indexing a lot for this both of this too, unless it gets very clunky with counting :. EDIT: Though I guess I admit that for squeezing, indexing does not give the information that there really is only one element.
There was a problem hiding this comment.
@charris: Added comments about reshape
eric-wieser
force-pushed
the
squeeze-expand-docs
branch
from
75d54a2
to
f2b27fb
Compare
|
Thanks Eric. |
[ci skip]'d
If #8735 might not be such a good idea, this gets most of the way there.