regression_modeler
Perform regression analysis: fit models (linear, polynomial, ridge, lasso, elastic net, logistic), predict outcomes, and compare model performance.
Instructions
Brief description: Regression analysis and machine learning modeling tool, supporting various regression algorithms and prediction functions.
Examples:
regression_modeler(operation='fit', x_data=[[1], [2], [3]], y_data=[2, 4, 6], model_type='linear')
regression_modeler(operation='predict', x_data=[[12]], training_x=[[1], [2], [3]], training_y=[2, 4, 6])
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| operation | No | Regression operation type. Supports: 'fit', 'predict', 'residual_analysis', 'model_comparison' | fit |
| x_data | No | Independent variable data as 2D list | |
| y_data | No | Dependent variable data as 1D list | |
| model_type | No | Regression model type. Supports: 'linear', 'polynomial', 'ridge', 'lasso', 'elastic_net', 'logistic' | linear |
| degree | No | Degree for polynomial regression | |
| alpha | No | Regularization parameter | |
| l1_ratio | No | Elastic Net L1 ratio | |
| cv_folds | No | Number of cross-validation folds | |
| test_size | No | Test set proportion | |
| y_true | No | True values for residual analysis | |
| y_pred | No | Predicted values for residual analysis | |
| models_results | No | List of model results for comparison | |
| training_x | No | Training independent variable data | |
| training_y | No | Training dependent variable data | |
| model_params | No | Pre-trained model parameters |