predictor.py 文件源码

def _print_ml_analytics_results_random_forest(self):
        try:
            final_model_obj = self.trained_final_model.named_steps['final_model']
        except:
            final_model_obj = self.trained_final_model

        print('\n\nHere are the results from our ' + final_model_obj.model_name)
        if self.name is not None:
            print(self.name)
        print('predicting ' + self.output_column)

        # XGB's Classifier has a proper .feature_importances_ property, while the XGBRegressor does not.
        if final_model_obj.model_name in ['XGBRegressor', 'XGBClassifier']:
            self._get_xgb_feat_importances(final_model_obj.model)

        else:
            trained_feature_names = self._get_trained_feature_names()

            try:
                trained_feature_importances = final_model_obj.model.feature_importances_
            except AttributeError as e:
                # There was a version of LightGBM that had this misnamed to miss the "s" at the end
                trained_feature_importances = final_model_obj.model.feature_importance_

            feature_infos = zip(trained_feature_names, trained_feature_importances)

            sorted_feature_infos = sorted(feature_infos, key=lambda x: x[1])

            print('Here are the feature_importances from the tree-based model:')
            print('The printed list will only contain at most the top 50 features.')
            for feature in sorted_feature_infos[-50:]:
                print(feature[0] + ': ' + str(round(feature[1], 4)))