Machine Learning Regression Classification Project

Current Implementation

• Linear Regression using RAPIDS cuML
• Custom Gradient Descent Implementation with GPU Acceleration
• Polynomial Feature Transformation
• Model Hyperparameter Tuning

Features

• GPU-accelerated model training using RAPIDS cuML
• GridSearchCV for optimal hyperparameter selection
• Pipeline implementation for data preprocessing and model training
• Custom gradient descent implementation with cupy for GPU optimization
• Model persistence and loading capabilities
• Comprehensive model evaluation metrics

Model Evaluation Metrics

• Mean Squared Error (MSE)
• Root Mean Squared Error (RMSE)
• Mean Absolute Error (MAE)
• R² Score
• Adjusted R² Score

Project Structure

LinearRegression/ ├── cuml_linear_regression.py # RAPIDS cuML implementation ├── gradient_descent.py # Custom gradient descent ├── test_model.py # Model evaluation ├── config.py # Configuration settings └── preprocess_df.py # Data preprocessing

Future Additions

• [To be added as the project evolves]