DOC Expand multilabel in decision function in glossary #24095
Conversation
There was a problem hiding this comment.
Thank you for the PR. It's interesting how the glossary goes straight to the decision function. When I introduce multi-label classification, I usually go to the target format first: https://scikit-learn.org/stable/modules/multiclass.html#multiclass-and-multioutput-algorithms
Overall, I think using a list makes the glossary entry clearer, so I am +1.
| interpreted, like in the binary case, by thresholding at 0. | ||
|
|
||
| TODO: `This gist | ||
| <https://gist.github.com/jnothman/4807b1b0266613c20ba4d1f88d0f8cf5>`_ |
There was a problem hiding this comment.
I think it is unclear what the TODO is here. @jnothman What did you have in mind?
Fair point, looking more closely I see that there is a target types section above: https://scikit-learn.org/stable/glossary.html#target-types - I've added a link to the glossary term 'multilabel' |
doc/glossary.rst
Outdated
|
|
||
| - Single 2d array of shape (`n_samples`, `n_labels`), with each | ||
| 'column' in the array corresponding to the individual binary | ||
| classification decisions. This is ambiguously identical to the |
There was a problem hiding this comment.
nitpick: Is "ambiguously" necessary?
|
LGTM Thanks @lucyleeow |
|
Thanks all! |
Reference Issues/PRs
(Continues from stalled PR) Closes #13660
closes #13533.
What does this implement/fix? Explain your changes.
Expand multilabel in decision function in glossary, using the suggestion: #13660 (comment)
I didn't include the code from the gist as no other section of the glossary included code. I could give an example based off of the code in the gist (e.g., dataset of 5 labels and 3 samples, the shape would be: ... ) but was not sure. Happy to take suggestions
Any other comments?