Change default parameters for feature selectors (#3110)

* use mean threshold only * docs * lint * Fix tests * Make default median and add to hyperparameter ranges
alteryx · Dec 2, 2021 · 36a50d2 · 36a50d2
1 parent 840fc3b
commit 36a50d2
Show file tree

Hide file tree

Showing 4 changed files with 15 additions and 8 deletions.
diff --git a/docs/source/release_notes.rst b/docs/source/release_notes.rst
@@ -4,6 +4,7 @@ Release Notes
     * Enhancements
         * Renamed ``DelayedFeatureTransformer`` to ``TimeSeriesFeaturizer`` and enhanced it to compute rolling features :pr:`3028`
     * Fixes
+        * Default parameters for ``RFRegressorSelectFromModel`` and ``RFClassifierSelectFromModel`` has been fixed to avoid selecting all features :pr:`3110`
     * Changes
     * Documentation Changes
     * Testing Changes

diff --git a/evalml/pipelines/components/transformers/feature_selection/rf_classifier_feature_selector.py b/evalml/pipelines/components/transformers/feature_selection/rf_classifier_feature_selector.py
@@ -1,5 +1,4 @@
 """Component that selects top features based on importance weights using a Random Forest classifier."""
-import numpy as np
 from sklearn.ensemble import RandomForestClassifier as SKRandomForestClassifier
 from sklearn.feature_selection import SelectFromModel as SkSelect
 from skopt.space import Real
@@ -29,11 +28,11 @@ class RFClassifierSelectFromModel(FeatureSelector):
     name = "RF Classifier Select From Model"
     hyperparameter_ranges = {
         "percent_features": Real(0.01, 1),
-        "threshold": ["mean", -np.inf],
+        "threshold": ["mean", "median"],
     }
     """{
         "percent_features": Real(0.01, 1),
-        "threshold": ["mean", -np.inf],
+        "threshold": ["mean", "median"],
     }"""
 
     def __init__(
@@ -42,7 +41,7 @@ def __init__(
         n_estimators=10,
         max_depth=None,
         percent_features=0.5,
-        threshold=-np.inf,
+        threshold="median",
         n_jobs=-1,
         random_seed=0,
         **kwargs,

diff --git a/evalml/pipelines/components/transformers/feature_selection/rf_regressor_feature_selector.py b/evalml/pipelines/components/transformers/feature_selection/rf_regressor_feature_selector.py
@@ -1,5 +1,4 @@
 """Component that selects top features based on importance weights using a Random Forest regresor."""
-import numpy as np
 from sklearn.ensemble import RandomForestRegressor as SKRandomForestRegressor
 from sklearn.feature_selection import SelectFromModel as SkSelect
 from skopt.space import Real
@@ -29,11 +28,11 @@ class RFRegressorSelectFromModel(FeatureSelector):
     name = "RF Regressor Select From Model"
     hyperparameter_ranges = {
         "percent_features": Real(0.01, 1),
-        "threshold": ["mean", -np.inf],
+        "threshold": ["mean", "median"],
     }
     """{
         "percent_features": Real(0.01, 1),
-        "threshold": ["mean", -np.inf],
+        "threshold": ["mean", "median"],
     }"""
 
     def __init__(
@@ -42,7 +41,7 @@ def __init__(
         n_estimators=10,
         max_depth=None,
         percent_features=0.5,
-        threshold=-np.inf,
+        threshold="median",
         n_jobs=-1,
         random_seed=0,
         **kwargs,

diff --git a/evalml/tests/component_tests/test_components.py b/evalml/tests/component_tests/test_components.py
@@ -888,6 +888,10 @@ def test_transformer_transform_output_type(X_y_binary):
                 component, SelectByType
             ):
                 assert transform_output.shape == (X.shape[0], 0)
+            elif isinstance(component, RFRegressorSelectFromModel):
+                assert transform_output.shape == (X.shape[0], 10)
+            elif isinstance(component, RFClassifierSelectFromModel):
+                assert transform_output.shape == (X.shape[0], 10)
             elif isinstance(component, PCA) or isinstance(
                 component, LinearDiscriminantAnalysis
             ):
@@ -915,6 +919,10 @@ def test_transformer_transform_output_type(X_y_binary):
                 component, SelectByType
             ):
                 assert transform_output.shape == (X.shape[0], 0)
+            elif isinstance(component, RFRegressorSelectFromModel):
+                assert transform_output.shape == (X.shape[0], 10)
+            elif isinstance(component, RFClassifierSelectFromModel):
+                assert transform_output.shape == (X.shape[0], 10)
             elif isinstance(component, PCA) or isinstance(
                 component, LinearDiscriminantAnalysis
             ):