sklearn-model-trainer

# Scikit-learn Model Trainer Train machine learning models using scikit-learn with cross-validation, hyperparameter tuning, and pipeline construction. ## Overview This skill provides comprehensive capabilities for training machine learning models using scikit-learn. It supports the full model development workflow from data preprocessing through model training, evaluation, and serialization. ## Capabilities ### Model Training - Train classification models (LogisticRegression, RandomForest, SVM, etc.) - Train regression models (LinearRegression, GradientBoosting, etc.) - Train clustering models (KMeans, DBSCAN, etc.) - Support for ensemble methods (VotingClassifier, Stacking, etc.) ### Cross-Validation - K-fold cross-validation - Stratified K-fold for imbalanced datasets - Time series split for temporal data - Leave-one-out and leave-p-out validation - Custom cross-validation strategies ### Hyperparameter Tuning - GridSearchCV for exhaustive search - RandomizedSearchCV for random sampling - Halving search strategies for efficiency - Custom scoring functions - Multi-metric evaluation ### Pipeline Construction - Feature preprocessing pipelines - Column transformers for heterogeneous data - Feature selection integration - Composite pipelines with caching ### Model Serialization - Save models with joblib (recommended) - Pickle serialization - ONNX export for interoperability - Model versioning support ## Prerequisites ### Installation ```bash pip install scikit-learn>=1....