datarobot-model-training

Automate DataRobot model training by coordinating project setup, target configuration, AutoML execution, and model comparison. This handles classification, regression, and time-series forecasting workflows using the DataRobot Python SDK and helper scripts to create projects, start training, and list models.

Start AutoML training by creating a project from your dataset, setting the target variable, and selecting a training mode like Quick or Comprehensive. The provided helper scripts and DataRobot Python SDK patterns guide you through configuring partitioning, executing training, and monitoring job status.

The DataRobot Python SDK supports time series forecasting alongside classification and regression tasks. You apply optional time-series settings within the project context to configure training options, monitor status, and select the best-performing forecasting models.

Compare models by interpreting key metrics and feature impact to choose a deployment-ready model. The workflow guides you through listing trained models and analyzing performance to identify the top-performing model for your specific classification or regression scenario.

You need an existing DataRobot project context before configuring training options and starting AutoML execution. The workflow requires this project context to set targets, apply optional time-series settings, and monitor training status to select the best-performing models.

DataRobot AutoML supports Quick, Comprehensive, and Manual training modes. You select the appropriate mode during configuration to start training, monitor status, and evaluate models based on key metrics and feature impact for your deployment-ready model selection.