Sklearn Cross Validation icon

Sklearn Cross Validation

Perform cross-validation on sklearn models

Sklearn Cross Validation

AI-generated

Summary

Run cross-validation on the chosen sklearn model, returning per-fold scores and metrics.

Inputs

  • modelType — Model type to use (e.g., LogisticRegression, RandomForestClassifier).
  • cvMethod — Cross-validation method (e.g., kfold, stratified).
  • folds — Number of folds for cross-validation.
  • featureColumns (required) — Comma-separated list of feature column names.
  • targetColumn (required) — Name of the target column.
  • scoring — Scoring metric for cross-validation.
  • randomState — Random seed for reproducibility.
  • pythonPath — Path to Python executable with scikit-learn installed.

Output shape

array of result objects

Each input row yields a result object with model_type, cv_method, scoring_metric, n_folds, mean_test_score, std_test_score, mean_train_score, std_train_score, test_scores_per_fold, train_scores_per_fold, fit_times, and total_samples; on error the object includes an 'error' field.

Examples

Example 1: Classify using Logistic Regression with 5-fold CV

modelType=LogisticRegression, cvMethod=kfold, folds=5, featureColumns=feature1,feature2, targetColumn=target, scoring=accuracy, randomState=42

Example 2: Regression with Random Forest Regressor, 3-fold CV

modelType=RandomForestRegressor, cvMethod=kfold, folds=3, featureColumns=age,income,feature3, targetColumn=salary, scoring=r2, randomState=123

Example 3: Classification with SVC and Stratified K-Fold

modelType=SVC, cvMethod=stratified, folds=10, featureColumns=age,income, targetColumn=churn, scoring=f1_weighted, randomState=7

Discussion