modeling

To train a machine learning model and tune hyperparameters, the Skill selects appropriate algorithms for your task and applies Optuna for parameter optimization, producing a saved model artifact. It handles data splitting, model fitting, and evaluation.

The best way to split data for regression and classification tasks is through Train/Validation/Test partitions, which this approach applies to ensure proper model fitting and accurate evaluation metrics before saving the trained artifacts.

Yes, you can use this for time-series forecasting alongside standard classification tasks. The Skill identifies and configures appropriate algorithms across regression, classification, and time-series data science projects.

To save trained model artifacts to a specified path, the Skill manages the output by concluding the data preparation, model fitting, and evaluation pipeline, storing the final trained model directly to your designated location.

Yes, the model training pipeline handles data preparation, model fitting, and evaluation. It automates these core stages, including data splitting and parameter tuning with Optuna, to build and train effective ML models.

Appropriate algorithms are selected based on your problem setting, spanning regression, classification, and time-series tasks. The Skill configures these algorithms within the training pipeline to match your specific dataset requirements.