Cross-validation in sklearn

• 📖 Cross-validation: evaluating estimator performance
• Cross-validation

Cross_val_score function

Say we trained a model using Support Vector Machine (SVM):

from sklearn.model_selection import cross_val_score
clf = svm.SVC(kernel='linear', C=1)
scores = cross_val_score(clf, X, y, cv=5)
print(scores)
 
print("%0.2f accuracy with a standard deviation of %0.2f" % (scores.mean(), scores.std()))

You can also choose another metric, for example:

scores = cross_val_score(clf,X,y,cv=5, scoring='f1_macro')

Cross_validate function

• 📖 Sklearn Docs - sklearn.model_selection.cross_validate This function is used when you want to specify multiple metrics for evaulation either/or you want more metadata like fit-times and score-times.

Multiple Metrics

scoring = ['precision_macro', 'recall_macro']
clf = svm.SVC(kernel='linear', C=1)
scores = cross_validate(clf, X, y, scoring=scoring)

Other techniques

There are a lot of techniques implemented in the sklearn library, you can read more about them here.