Performance Metrics

Create an array with model predictions y_pred by passing the test set X_test after the model has been trained, e.g.:

y_pred = model.predict(X_test)

This array is to be compared with the actual labels from the test set y_test.

Regression Performance Metrics

from sklearn.metrics import mean_absolute_error, mean_squared_error
mean_absolute_error(y_test,y_pred)
np.sqrt(mean_squared_error(y_test,y_pred))

Classification Performance Metrics

• Receiver operating characteristic ROC curve (only for binary classification):

from sklearn.metrics import roc_curve, RocCurveDisplay
fpr, tpr, thresholds = roc_curve(y_true,y_pred)
RocCurveDisplay(fpr=fpr, tpr=tpr).plot()

fpr and tpr stand for false and true positive rates.

• Confusion matrix:

from sklearn.metrics import confusion_matrix,ConfusionMatrixDisplay

cm = confusion_matrix(y_test,y_pred)
ConfusionMatrixDisplay(cm).plot()

• Classification report:

from sklearn.metrics import classification_report
print(classification_report(y_test,y_pred))

• Accuracy score:

from sklearn.metrics import accuracy_score
accuracy_score(y_test,y_pred)

• Other metrics for classification:

from sklearn.metrics import precision_score, recall_score

• F1 score = 2 * precision * recall / (precision + recall)