JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.preprocessing import StandardScaler

X_train = np.array([[2.0], [2.0], [6.0], [6.0]])
X_test = np.array([[10.0]])
scaler = StandardScaler()
scaler.fit(X_train)
z = scaler.transform(X_test)[0, 0]
print(round(z, 2))

3.0

Expecting 0.0: transform uses the mean 4.0 and standard deviation 2.0 stored by fit on X_train, so 10.0 becomes (10 - 4) / 2 = 3.0.
Expecting 6.0, the value minus the mean: standardization also divides by the standard deviation, which is 2.0 here.

What would scaler.fit_transform(X_test) give for the same test row, and why is that wrong?
Why may the scaled test set have a mean different from zero?

Back to questions Open topic in roadmap →Answer it yourself — Quiz

Back to questions

JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.preprocessing import StandardScaler

X_train = np.array([[2.0], [2.0], [6.0], [6.0]])
X_test = np.array([[10.0]])
scaler = StandardScaler()
scaler.fit(X_train)
z = scaler.transform(X_test)[0, 0]
print(round(z, 2))

3.0

Expecting 0.0: transform uses the mean 4.0 and standard deviation 2.0 stored by fit on X_train, so 10.0 becomes (10 - 4) / 2 = 3.0.
Expecting 6.0, the value minus the mean: standardization also divides by the standard deviation, which is 2.0 here.

What would scaler.fit_transform(X_test) give for the same test row, and why is that wrong?
Why may the scaled test set have a mean different from zero?

Back to questions Open topic in roadmap →Answer it yourself — Quiz