Maintenance code, support for sci kit-learn version 1.2.2.

README.md

@@ -104,7 +104,8 @@ to run stability selection with complementary pairs bootstrapping.
 
 ## Feedback and contributing
 
-Feedback and contributions are much appreciated. If you have any feedback, please post it on the [issue tracker](https://github.com/scikit-learn-contrib/stability-selection/issues). 
+* Feedback and contributions are much appreciated. If you have any feedback, please post it on the [issue tracker](https://github.com/scikit-learn-contrib/stability-selection/issues). 
+
 
 ## References
 
@@ -115,3 +116,4 @@ Feedback and contributions are much appreciated. If you have any feedback, pleas
    error control: another look at stability selection. Journal
    of the Royal Statistical Society: Series B (Statistical Methodology),
     75(1), pp.55-80.
+[3] https://github.com/scikit-learn-contrib/stability-selection

examples/plot_randomized_lasso_path.py

@@ -40,9 +40,9 @@ def generate_experiment_data(n=200, p=200, rho=0.6, random_state=3245):
     lambda_grid = np.linspace(0.001, 0.5, num=100)
 
     for weakness in [0.2, 0.5, 1.0]:
-        estimator = RandomizedLasso(weakness=weakness)
-        selector = StabilitySelection(base_estimator=estimator, lambda_name='alpha',
-                                      lambda_grid=lambda_grid, threshold=0.9, verbose=1)
+        estimator = RandomizedLasso(weakness=weakness,normalize=True)
+        selector = StabilitySelection(base_estimator=estimator, lambda_name='alpha',lambda_grid=lambda_grid,
+                                      threshold=0.9, verbose=1)
         selector.fit(X, y)
 
         fig, ax = plot_stability_path(selector)

examples/plot_stability_scores.py

@@ -36,10 +36,9 @@ def _generate_dummy_classification_data(p=1000, n=1000, k=5, random_state=123321
 
     base_estimator = Pipeline([
         ('scaler', StandardScaler()),
-        ('model', LogisticRegression(penalty='l1'))
+        ('model', LogisticRegression())
     ])
-    selector = StabilitySelection(base_estimator=base_estimator, lambda_name='model__C',
-                                  lambda_grid=np.logspace(-5, -1, 50))
+    selector = StabilitySelection(base_estimator=base_estimator, lambda_name='model__C',lambda_grid=np.logspace(-5, -1, 50))
     selector.fit(X, y)
 
     fig, ax = plot_stability_path(selector)

requirements.txt

@@ -1,4 +1,4 @@
-nose>=1.1.2
-scikit-learn>=0.19
-matplotlib>=2.0.0
-numpy>=1.8.0
+nose==1.3.7
+scikit-learn>=1.2.2
+matplotlib==3.7.1
+numpy==1.24.3

stability_selection/randomized_lasso.py

@@ -18,7 +18,7 @@
 from scipy.sparse import issparse
 
 from sklearn.linear_model import LogisticRegression, Lasso
-from sklearn.linear_model.base import _preprocess_data
+from sklearn.linear_model._base import _preprocess_data
 from sklearn.utils import check_X_y, check_random_state
 
 __all__ = ['RandomizedLogisticRegression', 'RandomizedLasso']
@@ -120,9 +120,9 @@ def __init__(self, weakness=0.5, alpha=1.0, fit_intercept=True, normalize=False,
                  tol=1e-4, warm_start=False, positive=False,
                  random_state=None, selection='cyclic'):
         self.weakness = weakness
+        self.normalize = normalize
         super(RandomizedLasso, self).__init__(
-            alpha=alpha, fit_intercept=fit_intercept,
-            normalize=normalize, precompute=precompute, copy_X=copy_X,
+            alpha=alpha, fit_intercept=fit_intercept, precompute=precompute, copy_X=copy_X,
             max_iter=max_iter, tol=tol, warm_start=warm_start,
             positive=positive, random_state=random_state,
             selection=selection)
@@ -149,9 +149,10 @@ def fit(self, X, y):
 
         weights = weakness * random_state.randint(0, 1 + 1, size=(n_features,))
 
-        # TODO: I am afraid this will do double normalization if set to true
-        #X, y, _, _ = _preprocess_data(X, y, self.fit_intercept, normalize=self.normalize, copy=False,
-        #             sample_weight=None, return_mean=False)
+        # TODO: I am afraid this will do double normalization if set to true.
+        X, y, X_offset, y_offset, X_scale = _preprocess_data(X, y, self.fit_intercept,
+                                                             normalize=self.normalize,copy=False,
+                                                             sample_weight=None,check_input=True)
 
         # TODO: Check if this is a problem if it happens before standardization
         X_rescaled = _rescale_data(X, weights)

stability_selection/stability_selection.py

@@ -19,11 +19,12 @@
 """
 
 from warnings import warn
-
 import matplotlib.pyplot as plt
 import numpy as np
 from sklearn.base import BaseEstimator, TransformerMixin, clone
-from sklearn.externals.joblib import Parallel, delayed
+from joblib import Parallel, delayed
+
+
 from sklearn.feature_selection import SelectFromModel
 from sklearn.linear_model import LogisticRegression
 from sklearn.pipeline import Pipeline