Sklearn Feature Selection icon

Sklearn Feature Selection

Select best features using various sklearn methods

Sklearn Feature Selection

AI-generated

Summary

Select best features from input data using scikit-learn feature selection methods.

Inputs

  • method — Feature selection method to use (selectKBest, selectPercentile, varianceThreshold, rfe).
  • featureColumns (required) — Comma‑separated list of feature column names.
  • targetColumn (required) — Name of the target column.
  • k — Number of top features to select (used with selectKBest or rfe).
  • percentile — Percent of features to keep (0‑100) (used with selectPercentile).
  • threshold — Variance threshold below which features are removed (used with varianceThreshold).
  • scoreFunc — Score function for selectKBest or selectPercentile.
  • estimator — Base estimator for RFE.
  • pythonPath — Path to the Python executable with scikit-learn installed.

Output shape

list of items containing transformed data and feature selection metadata

each item includes 'selected_features' and the transformed feature array; optional fields may be present (feature_scores, feature_variances, feature_rankings). On error an 'error' field may be returned.

Examples

Example 1: Select top 5 classification features using ANOVA F-score

method=selectKBest, k=5, scoreFunc=f_classif, featureColumns=age,income,education, targetColumn=label

Example 2: Perform Recursive Feature Elimination with a Random Forest classifier

method=rfe, estimator=RandomForestClassifier, k=10, featureColumns=feature1,feature2,feature3, targetColumn=target

Discussion