medium10 pts · first answer countsml-pipelinesdata-leakagescikit-learn

Scaling before splitting

What is wrong with this preprocessing order?

X_scaled = StandardScaler().fit_transform(X)
X_train, X_test, y_train, y_test = train_test_split(X_scaled, y)

A.StandardScaler can never be used before a splitB.train_test_split will fail on any scaled dataC.Nothing is wrong; the order of steps does not matterD.The scaler sees test rows, so test scores are optimistic

Discussion 0

Join free to post in the discussion.

Scaling before splitting

Discussion 0

Related questions