Classification

Classification Skill

This skill handles supervised classification tasks using scikit-learn.

Capabilities

• Binary and multi-class classification
• Algorithm selection (LogisticRegression, RandomForest, GradientBoosting, SVM, KNN)
• Preprocessing pipelines (scaling, encoding, imputation)
• Cross-validation and hyperparameter tuning (GridSearchCV, RandomizedSearchCV)
• Metric evaluation (accuracy, precision, recall, F1, ROC-AUC, confusion matrix)
• Class imbalance handling (stratified splits, class weights, SMOTE)

When to Use

Use this skill when the objective involves predicting a categorical label from input features. Examples:

• "Classify iris species from petal measurements"
• "Predict customer churn (yes/no)"
• "Categorize emails as spam or not spam"
• "Detect fraudulent transactions"

Approach