[MRG] DOC update UG and docstrings for isotonic regression#16234
[MRG] DOC update UG and docstrings for isotonic regression#16234agramfort merged 10 commits intoscikit-learn:masterfrom
Conversation
adrinjalali
left a comment
There was a problem hiding this comment.
Haven't checked the math, otherwise there are some still left in the formattings which need to be fixed, the T_ on the transform method for instance. Sorry I meant to get to this, you got there faster.
sklearn/isotonic.py
Outdated
| The string value "auto" determines whether y should | ||
| increase or decrease based on the Spearman correlation estimate's | ||
| sign. | ||
| increasing : bool or 'auto', default= True |
There was a problem hiding this comment.
| increasing : bool or 'auto', default= True | |
| increasing : bool or 'auto', default=True |
We cannot change that though, can we? The current signature is |
thomasjpfan
left a comment
There was a problem hiding this comment.
Do we talk about increasing='auto' in the user guide?
Co-Authored-By: Thomas J Fan <thomasjpfan@gmail.com>
|
Closes #16329 Comments were addressed, does this have your +1 @adrinjalali @thomasjpfan ? |
|
There are still some left in the other places in the file, for instance: I think last time I checked there were a few ones which needed to be fixed |
allefeld
left a comment
There was a problem hiding this comment.
Looks good generally, minor comments.
sklearn/isotonic.py
Outdated
| y_min : float, default=None | ||
| If not None, set the lowest value of the fit to y_min. | ||
| Lower bound on the lowest predicted value (the minimum value may | ||
| still be higher). Defaults to -inf. |
There was a problem hiding this comment.
I think it might be confusing that it first says default=None, but then Defaults to -inf. How about "If not None, lower bound ..." or "If specified, lower bound ..."
sklearn/isotonic.py
Outdated
| Returns | ||
| ------- | ||
| T_ : array, shape=(n_samples,) | ||
| y_pred : ndarry of shape (n_samples,) |
|
First time I did a review on GH – I chose "approve" though I had comments, not sure whether that's ok? |
|
Thanks @allefeld !
Sure. I personally press approve when comments are minor, or when I know they'll be easily addressed so I don't need to check back |
…arn#16234) * Isotonic regression docs * remove space * Apply suggestions from code review Co-Authored-By: Thomas J Fan <thomasjpfan@gmail.com> * Documented auto option in UG * fixed more docstrings * Addressed comments Co-authored-by: Thomas J Fan <thomasjpfan@gmail.com>
…arn#16234) * Isotonic regression docs * remove space * Apply suggestions from code review Co-Authored-By: Thomas J Fan <thomasjpfan@gmail.com> * Documented auto option in UG * fixed more docstrings * Addressed comments Co-authored-by: Thomas J Fan <thomasjpfan@gmail.com>
also removed math equations from docstrings (some are outdated, some are wrong)