Signature
RocCurveDisplay.from_estimator(estimator, X, y)What it does
RocCurveDisplay.from_estimator scores the model, computes false and true positive rates, and plots the curve. from_predictions plots precomputed probabilities instead.
Parameters
estimatorestimator- A fitted classifier with predict_proba or decision_function.
Xarray-like- Feature matrix to score.
yarray-like- True binary labels.
Returns
RocCurveDisplay
Example
pythonRocCurveDisplay example
import matplotlib.pyplot as plt
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import RocCurveDisplay
model = LogisticRegression().fit(X_train, y_train)
RocCurveDisplay.from_estimator(model, X_test, y_test)
plt.show()Watch out for
- A curve computed on training data looks optimistic; plot the test fold.
- ROC can look strong when positives are rare; read the precision-recall curve too.