How to get cross validation scores from AutoML object?

[stepthom]

trafficstars

How can I obtain the cross validation scores that were obtained during the call to fit()?

E.g.:

 from supervised.automl import AutoML # mljar-supervised
        
        automl_settings = {
            "mode": "Compete",
            "model_time_limit": search_time,
            "eval_metric": eval_metric,
            "stack_models": True,
            "explain_level": 0,
            "feature_selection": False,
            "algorithms": ["Linear", "Decision Tree", "Random Forest", "LightGBM", "Xgboost", "Catboost", "Neural Network"],
            "golden_features": True,
            "n_jobs": 8,
            "verbose": 1,
        }
        
        pipe = AutoML(**automl_settings)
        pipe.fit(X_train, y_train)

       # TODO: what to put here?
       # pipe.scores_  ???

[pplonski]

There should be output directory with all results created after fit. There will be readme file with markdown report. If you are in notebook then you can also call pipe.report() to display interctive report with all results.

[stepthom]

Thanks, @pplonski. In the future, it might make sense to add a feature to allow these scores to be obtained via the AuotML object directly so that one can get them programmatically after fit().

[pplonski]

There is one more option pipe.get_leaderbord(). Does it work for you?