def test_final_model(self, X, y): """ Test the (already trained) model pipeline on the provided test data (X and y). Store the test judgment metric and return the rest of the metrics as a hierarchical dictionary. """ # time the prediction start_time = time.time() total = time.time() - start_time self.avg_predict_time = old_div(total, float(len(y))) # TODO: this is hacky. See https://github.com/HDI-Project/ATM/issues/48 binary = self.num_classes == 2 kwargs = {} if self.verbose_metrics: kwargs['include_curves'] = True if not binary: kwargs['include_per_class'] = True # compute the actual test scores! test_scores = test_pipeline(self.pipeline, X, y, binary, **kwargs) # save meta-metrics self.test_judgment_metric = test_scores.get(self.judgment_metric) return test_scores
def test_final_model(self, X, y): """ Test the (already trained) model pipeline on the provided test data (X and y). Store the test judgment metric and return the rest of the metrics as a hierarchical dictionary. """ # time the prediction starttime = time.time() y_preds = self.pipeline.predict(X) binary = self.num_classes == 2 test_scores = test_pipeline(self.pipeline, X, y, binary) total = time.time() - starttime self.avg_prediction_time = total / float(len(y)) self.test_judgment_metric = test_scores.get(self.judgment_metric) return test_scores