models/supervised/classification/README.md

Classification Models

This directory contains Python scripts that define various classification models and their associated hyperparameter grids. Each model file sets up a scikit-learn-compatible estimator and defines a parameter grid for use with the train_classification_model.py script.

These model definition files:

• Specify an estimator (e.g., LogisticRegression(), RandomForestClassifier(), XGBClassifier()).
• Define a param_grid dict for hyperparameter tuning using GridSearchCV.
• Optionally provide a default_scoring metric (e.g., accuracy).
• Work for both binary and multi-class classification tasks.
• Are intended to be flexible and modular, allowing easy swapping of models without changing other parts of the code.

Note: Preprocessing steps, hyperparameter tuning logic, and label encoding for categorical targets are handled externally by the scripts and utilities.

Available Classification Models

• Logistic Regression
• Decision Tree Classifier
• Random Forest Classifier (Bagging)
• Extra Trees Classifier
• Gradient Boosting Classifier (Boosting)
• AdaBoost Classifier (Boosting)
• XGBoost Classifier (Boosting)
• LightGBM Classifier (Boosting)
• CatBoost Classifier (Boosting)
• Support Vector Classifier (SVC)
• K-Nearest Neighbors (KNN) Classifier
• Multilayer Perceptron (MLP) Classifier
• GaussianNB (Naive Bayes Classifier)
• Linear Discriminant Analysis (LDA)
• Quadratic Discriminant Analysis (QDA)

To train any of these models, specify the --model_module argument with the appropriate model name (e.g., logistic_regression) when running train_classification_model.py.