New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
DOC Adds release highlights for 1.2 #24798
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am fine with merging this as it is to get it started and allow follow-up PRs to improve and extend this doc.
|
||
X_out = preprocessor.fit_transform(X) | ||
X_out.sample(n=5, random_state=0) | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What about linking to or even embedding @amueller's video here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That makes sense. I suspect there is a privacy concern with embedding a Youtube video.
I added a link to Andy's video to the highlights.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@thomasjpfan
Can we link to this video instead? We updated it so the sound was easier to hear:
https://youtu.be/5bCg8VfX2x8
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated link in 06c065b
(#24798)
|
||
_, ax = plt.subplots(figsize=(12, 4)) | ||
PartialDependenceDisplay.from_estimator(hist, X, ["bp", "bmi"], n_cols=2, ax=ax) | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is not easy to interpret this plot without additional explanations.
Wouldn't it make sense to also plot the PDB for the same model without the interaction_cst
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think adding an explanation would detract from the highlights. With that in mind, I removed the partial dependence curve. If a reader wants to learn more, there is a link to the User Guide.
REF: The highlights for missing value support in HistGradientBoosting*
only shows that fit
works.
Co-authored-by: Adrin Jalali <adrin.jalali@gmail.com>
@@ -27,7 +27,7 @@ | |||
# scikit-learn's transformers now support pandas output with the `set_output` API. | |||
# To learn more about the `set_output` API see the example: | |||
# :ref:`sphx_glr_auto_examples_miscellaneous_plot_set_output.py` and | |||
# this `YouTube video <https://www.youtube.com/watch?v=J4KCu9WWDTo>`__. | |||
# this `YouTube video <https://youtu.be/5bCg8VfX2x8>`__. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@thomasjpfan Seems like I don't have rights to suggest edits to this PR.
In any case, can we change:
# this `YouTube video <https://youtu.be/5bCg8VfX2x8>`__.
to
# this `video, pandas DataFrame output for scikit-learn transformers (some examples) <https://youtu.be/5bCg8VfX2x8>`__.
See note on accessible hyperlinks: data-umbrella/dasch#6 (comment)
X, y = load_diabetes(return_X_y=True, as_frame=True) | ||
|
||
hist_no_interact = HistGradientBoostingRegressor( | ||
interaction_cst=[[i] for i in range(X.shape[1])], random_state=0 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Once #24849 is merged we can update this to use the shortcut
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM.
Thanks @thomasjpfan |
Reference Issues/PRs
Towards #24664
What does this implement/fix? Explain your changes.
This PR adds release highlights for
set_output
, interaction constraints in HistGradientBoosting, and Array API.Any other comments?
As we did with the release notes for 1.1, we can have follow up PRs for additional items.