How to save the LightGBM Regressor trained using ATOM Pipeline to get prediction intervals using MAPIE

[ashish-wai]

Contribution guidelines

• [X] I've read the contribution guidelines.
• [X] The documentation does not mention anything about this feature.
• [X] There are no open or closed issues that are related to this feature.

Description

Hi, I read that we can use joblib to dump the entire pipeline created using ATOM-ML in an existing issue. How can I save only the trained model (XGB or LGB) to use MAPIE to get a prediction interval for regression rather than a single-point prediction?

https://mapie.readthedocs.io/en/latest/, or other libraries for other tasks (like explainability, etc).

Use Cases

• Conformal Regression
• Feature Importance
• Interoperability of the trained models

Screenshots / Mockups

No response

[tvdboom]

You can use the save_estimator method of the LightGBM model, e.g., atom.lgb.save_estimator("lightgbm")