From d87389a135709ce28c0231918de8b63cd4ca93fb Mon Sep 17 00:00:00 2001 From: Mark Aron Szulyovszky Date: Thu, 6 Jan 2022 18:13:19 +0100 Subject: [PATCH] fix(Selection): use step size of 5, speeding up feature selection by 5x (overall 3x improvement) (#115) --- feature_selection/feature_selection.py | 2 +- models/model_map.py | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/feature_selection/feature_selection.py b/feature_selection/feature_selection.py index d0e7685..2b46d5e 100644 --- a/feature_selection/feature_selection.py +++ b/feature_selection/feature_selection.py @@ -33,7 +33,7 @@ def __select_features(X: pd.DataFrame, y: pd.Series, model: Model, n_features_to feat_selector_model = backup_model.model # selector = RFECV(feat_selector_model, cv = cv, step=5, min_features_to_select=min_features_to_select) - selector = RFE(feat_selector_model, n_features_to_select= n_features_to_select) + selector = RFE(feat_selector_model, n_features_to_select= n_features_to_select, step=5) selector = selector.fit(X_scaled, y) print("Kept %d features out of %d" % (selector.n_features_, X_scaled.shape[1])) diff --git a/models/model_map.py b/models/model_map.py index cc93502..b4ea05f 100644 --- a/models/model_map.py +++ b/models/model_map.py @@ -1,4 +1,5 @@ -from sklearn.linear_model import LinearRegression, Lasso, BayesianRidge, LogisticRegression, Ridge +from sklearn.linear_model import LinearRegression, Lasso, BayesianRidge, Ridge +from sklearnex.linear_model import LogisticRegression from sklearn.tree import DecisionTreeClassifier from sklearnex.neighbors import KNeighborsRegressor, KNeighborsClassifier from sklearn.discriminant_analysis import LinearDiscriminantAnalysis