Guillaume Lemaitre

MAINT define a single time _estimator_has and refactor code

2

From past discussion, I realized that we are defining the same `_estimator_has` in several places while it does exactly the same job and has the same semantic. I think we...

MAINT add parameter validation using BaseEstimator

5

This bring parameter validation that is included in the supported scikit-learn version.

[RFC] mRMR feature selection

17

Speaking with @jorisvandenbossche IRL, we come to discuss about the mRMR feature selection among other methods. #5372 intended at first to implement the mRMR with mutual information as a metric....

This is reworking the first section to bring more into light the `tabular_learner` before the `TableVectorizer`.

This SLEP intends to have a discussion to define a clear guideline allowing to accept external contribution in the scikit-learn blog post.

SimpleImputer does not drop a column full of `np.nan` even when `keep_empty_feature=False`

5

The following code snippet lead to some surprises: ```python import numpy as np from sklearn.datasets import load_iris from sklearn.impute import SimpleImputer X, y = load_iris(return_X_y=True) X[:, 0] = np.nan imputer...

DOC update FAQ regarding dataframe I/O and support for categorical variable

2

closes #28930 Updating the FAQ to better reflect the current state of support for: - categorical variables in estimators - dataframes input/output

Just a retry.

ENH expose `max_memory_mb` parameter to contol trade-off speed/memory during computation

1

Related to https://github.com/scikit-learn/scikit-learn/pull/28375 NB: This PR is building upon https://github.com/scikit-learn/scikit-learn/pull/30104 that simplify a bit the code. TODO: - [x] changelog entry - [x] add tests This PR proposes to add...

DOC remove redundant example multiclass logistic regression

2

Partially addressing #27151 This PR removes one of the example of multiclass logistic regression. In addition, it improves the example that we keep and discuss in more details the difference...