adding versionadded to a few modules#16015
adding versionadded to a few modules#16015adrinjalali wants to merge 53 commits intoscikit-learn:masterfrom
Conversation
Tversions
This reverts commit ef7d51f.
This reverts commit 56226ec.
…it-learn into version_docstrings
Tversions
Version docstrings
…roupOut, LeavePGroupsOut: Issue scikit-learn#15426
…rn into adding_version Merging my work with the other group's so that our versionadded labels all get added at the same time.
…into adding_version
…into adding_version
jnothman
left a comment
There was a problem hiding this comment.
I still might be more comfortable if this was implemented one what's new file at a time (and certainly not with the 2 files max limit proposed on the other PR). and with care to only include newly introduced estimators...
| If a callable is passed it is used to extract the sequence of features | ||
| out of the raw, unprocessed input. | ||
|
|
||
| .. versionchanged:: 0.21 |
There was a problem hiding this comment.
I think this requires same formatting as above
| Xin Dang, Hanxiang Peng, Xueqin Wang and Heping Zhang | ||
| http://home.olemiss.edu/~xdang/papers/MTSE.pdf | ||
|
|
||
| .. versionadded:: v0.16 |
|
|
||
| Read more in the :ref:`User Guide <cross_validation>`. | ||
|
|
||
| .. versionadded:: 0.13 |
There was a problem hiding this comment.
depends how you account for the transition from sklearn.cross_validation to sklearn.model_selection...
| with a test set of size ``n_samples//(n_splits + 1)``, | ||
| where ``n_samples`` is the number of samples. | ||
|
|
||
| ..versionadded:: v0.18 |
There was a problem hiding this comment.
We have 0.19 in the summary above... and 0.18 here in the Notes
| Information Retrieval. Cambridge University Press, pp. 234-265. | ||
| https://nlp.stanford.edu/IR-book/html/htmledition/naive-bayes-text-classification-1.html | ||
|
|
||
| .. versionchanged:: v0.16 |
There was a problem hiding this comment.
This can't be in References. and I'm not sure whether we want this level of detail, if it implies maintaining such per-estimator changelogs elsewhere (which would be neat but is a lot of work)
| (Vol. 3, pp. 616-623). | ||
| https://people.csail.mit.edu/jrennie/papers/icml03-nb.pdf | ||
|
|
||
| .. versionadded:: v0.20 |
There was a problem hiding this comment.
In references, and includes "v"
|
|
||
| Read more in the :ref:`User Guide <classification>`. | ||
|
|
||
| .. versionadded:: 0.22 |
|
|
||
| Read more in the :ref:`User Guide <classification>`. | ||
|
|
||
| .. versionadded:: 0.14 |
| foo_param | ||
| expected_fit_params | ||
|
|
||
| .. versionchanged:: 0.22 |
There was a problem hiding this comment.
Not in Parameters... probably can rm altogether
@jnothman I'm happy to close this and have one changelog at a time as you propose, which should be fine as a few issues for the sprints. But you also leave a review, which makes me think I should apply them :P should I ignore your review and close this and the other PR and leave it as you suggest for the sprints? |
|
That depends on how much you trust the numbers in this PR, which I have not
checked all of. If you don't think they can be trusted, have someone start
again.
|
|
@adrinjalali, @jnothman , may I suggest to close this PR and let the exercise of checking the version_added to people at the sprint? @glemaitre has added some recommendations about how to make the PR more reviewable. Feel free to complete if necessary. |
|
Yep, closing. Thank you all :) |
Closes #15973
Somehow I couldn't push to the original repo, but had addressed some of the comments.