feature tags (#625)

koaning · Mar 19, 2024 · 7a1e283 · 7a1e283
1 parent 47c29a5
commit 7a1e283
Show file tree

Hide file tree

Showing 5 changed files with 9 additions and 5 deletions.
diff --git a/docs/user-guide/feature-selection.md b/docs/user-guide/feature-selection.md
@@ -2,6 +2,8 @@
 
 ## Maximum Relevance Minimum Redundancy
 
+!!! info "New in version 0.8.0"
+
 The [`Maximum Relevance Minimum Redundancy`][MaximumRelevanceMinimumRedundancy-api] (MRMR) is an iterative feature selection method commonly used in data science to select a subset of features from a larger feature set. The goal of MRMR is to choose features that have high *relevance* to the target variable while minimizing *redundancy* among the already selected features.
 
 MRMR is heavily dependent on the two functions used to determine relevace and redundancy. However, the paper [Maximum Relevanceand Minimum Redundancy Feature Selection Methods for a Marketing Machine Learning Platform](https://arxiv.org/pdf/1908.05376.pdf) shows that using [f_classif](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.f_classif.html) or [f_regression](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.f_regression.html) as relevance function and Pearson correlation as redundancy function is the best choice for a variety of different problems and in general is a good choice.
@@ -57,7 +59,7 @@ Feature selection method: mrmr_smile
 F1 score: 0.849
 ```
 
-The MRMR feature selection model provides better results compared against the other methods, although the smile technique performs rather good as well. 
+The MRMR feature selection model provides better results compared against the other methods, although the smile technique performs rather good as well.
 
 Finally, we can take a look at the selected features.
 

diff --git a/docs/user-guide/meta-models.md b/docs/user-guide/meta-models.md
@@ -136,7 +136,7 @@ Note that these predictions seems to yield the lowest error but take it with a g
 
 ### Specialized Estimators
 
-!!! info "New in version 0.7.5"
+!!! info "New in version 0.8.0"
 
 Instead of using the generic `GroupedPredictor` directly, it is possible to work with _task specific_ estimators, namely: [`GroupedClassifier`][grouped-classifier-api] and [`GroupedRegressor`][grouped-regressor-api].
 

diff --git a/pyproject.toml b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "scikit-lego"
-version = "0.7.4"
+version = "0.8.0"
 description="A collection of lego bricks for scikit-learn pipelines"
 
 license = {file = "LICENSE"}

diff --git a/sklego/feature_selection/mrmr.py b/sklego/feature_selection/mrmr.py
@@ -83,6 +83,8 @@ class MaximumRelevanceMinimumRedundancy(SelectorMixin, BaseEstimator):
 
             - np.ndarray, shape = (len(left), ), The array containing the redundancy score using the custom function.
 
+    !!! info "New in version 0.8.0"
+
     Parameters
     ----------
     k : int

diff --git a/sklego/meta/grouped_predictor.py b/sklego/meta/grouped_predictor.py
@@ -397,7 +397,7 @@ class GroupedRegressor(GroupedPredictor, RegressorMixin):
     Its spec is the same as [`GroupedPredictor`][sklego.meta.grouped_predictor.GroupedPredictor] but it is available
     only for regression models.
 
-    !!! info "New in version 0.7.5"
+    !!! info "New in version 0.8.0"
     """
 
     def fit(self, X, y):
@@ -434,7 +434,7 @@ class GroupedClassifier(GroupedPredictor, ClassifierMixin):
     Its equivalent to [`GroupedPredictor`][sklego.meta.grouped_predictor.GroupedPredictor] with `shrinkage=None`
     but it is available only for classification models.
 
-    !!! info "New in version 0.7.5"
+    !!! info "New in version 0.8.0"
     """
 
     def __init__(