Regression

Regression Skill

This skill handles supervised regression tasks using scikit-learn.

Capabilities

• Linear and non-linear regression
• Algorithm selection (LinearRegression, Ridge, Lasso, ElasticNet, RandomForestRegressor, GradientBoostingRegressor, SVR)
• Preprocessing pipelines (scaling, encoding, imputation)
• Cross-validation and hyperparameter tuning
• Metric evaluation (MAE, RMSE, R-squared, explained variance)
• Feature importance and partial dependence analysis

When to Use

Use this skill when the objective involves predicting a continuous numeric value. Examples:

• "Predict house prices from features"
• "Estimate customer lifetime value"
• "Forecast sales revenue"
• "Predict temperature from sensor readings"

Approach

1. Load and inspect the dataset (automated EDA)