From 0986e37b1202bf2790b911c8fa8b91c71d8eaed4 Mon Sep 17 00:00:00 2001 From: Rahil Parikh Date: Wed, 12 Apr 2023 13:35:02 +0530 Subject: [PATCH 1/2] roc_auc_score uses y_prob instead of y_pred --- examples/calibration/plot_calibration_curve.py | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/examples/calibration/plot_calibration_curve.py b/examples/calibration/plot_calibration_curve.py index 5f8ad621bc7a8..bc4772a326ce2 100644 --- a/examples/calibration/plot_calibration_curve.py +++ b/examples/calibration/plot_calibration_curve.py @@ -155,15 +155,18 @@ y_pred = clf.predict(X_test) scores["Classifier"].append(name) - for metric in [brier_score_loss, log_loss]: + for metric in [brier_score_loss, log_loss, roc_auc_score]: score_name = metric.__name__.replace("_", " ").replace("score", "").capitalize() scores[score_name].append(metric(y_test, y_prob[:, 1])) - for metric in [precision_score, recall_score, f1_score, roc_auc_score]: + for metric in [precision_score, recall_score, f1_score]: score_name = metric.__name__.replace("_", " ").replace("score", "").capitalize() scores[score_name].append(metric(y_test, y_pred)) score_df = pd.DataFrame(scores).set_index("Classifier") + score_df = score_df[ + ["Brier loss", "Log loss", "Precision ", "Recall ", "F1 ", "Roc auc "] + ] score_df.round(decimals=3) score_df @@ -300,15 +303,18 @@ def predict_proba(self, X): y_pred = clf.predict(X_test) scores["Classifier"].append(name) - for metric in [brier_score_loss, log_loss]: + for metric in [brier_score_loss, log_loss, roc_auc_score]: score_name = metric.__name__.replace("_", " ").replace("score", "").capitalize() scores[score_name].append(metric(y_test, y_prob[:, 1])) - for metric in [precision_score, recall_score, f1_score, roc_auc_score]: + for metric in [precision_score, recall_score, f1_score]: score_name = metric.__name__.replace("_", " ").replace("score", "").capitalize() scores[score_name].append(metric(y_test, y_pred)) score_df = pd.DataFrame(scores).set_index("Classifier") + score_df = score_df[ + ["Brier loss", "Log loss", "Precision ", "Recall ", "F1 ", "Roc auc "] + ] score_df.round(decimals=3) score_df From 7e511546492bf8d7f07727dfb197ee68dd98f24b Mon Sep 17 00:00:00 2001 From: Rahil Parikh Date: Thu, 13 Apr 2023 17:39:56 +0530 Subject: [PATCH 2/2] changes --- examples/calibration/plot_calibration_curve.py | 6 ------ 1 file changed, 6 deletions(-) diff --git a/examples/calibration/plot_calibration_curve.py b/examples/calibration/plot_calibration_curve.py index bc4772a326ce2..82b054aea4901 100644 --- a/examples/calibration/plot_calibration_curve.py +++ b/examples/calibration/plot_calibration_curve.py @@ -164,9 +164,6 @@ scores[score_name].append(metric(y_test, y_pred)) score_df = pd.DataFrame(scores).set_index("Classifier") - score_df = score_df[ - ["Brier loss", "Log loss", "Precision ", "Recall ", "F1 ", "Roc auc "] - ] score_df.round(decimals=3) score_df @@ -312,9 +309,6 @@ def predict_proba(self, X): scores[score_name].append(metric(y_test, y_pred)) score_df = pd.DataFrame(scores).set_index("Classifier") - score_df = score_df[ - ["Brier loss", "Log loss", "Precision ", "Recall ", "F1 ", "Roc auc "] - ] score_df.round(decimals=3) score_df