JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.ensemble import HistGradientBoostingClassifier

X = np.array([[1.0], [np.nan], [3.0], [4.0]] * 10)
y = np.array([0, 0, 1, 1] * 10)
model = HistGradientBoostingClassifier(random_state=0).fit(X, y)
print(model.predict(np.array([[np.nan]]))[0])

Expecting ValueError, as from LogisticRegression: HistGradientBoostingClassifier accepts np.nan and learns at each split which side the rows with a missing value go to.
Expecting nan: predict returns a class label, and a sample with np.nan follows the side its training rows took, here together with x = 1.0 and class 0.

Where would a sample with np.nan go if the training data had no missing values in this feature?
When is filling the gaps with SimpleImputer still worth doing before boosting?

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.ensemble import HistGradientBoostingClassifier

X = np.array([[1.0], [np.nan], [3.0], [4.0]] * 10)
y = np.array([0, 0, 1, 1] * 10)
model = HistGradientBoostingClassifier(random_state=0).fit(X, y)
print(model.predict(np.array([[np.nan]]))[0])

Expecting ValueError, as from LogisticRegression: HistGradientBoostingClassifier accepts np.nan and learns at each split which side the rows with a missing value go to.
Expecting nan: predict returns a class label, and a sample with np.nan follows the side its training rows took, here together with x = 1.0 and class 0.

Where would a sample with np.nan go if the training data had no missing values in this feature?
When is filling the gaps with SimpleImputer still worth doing before boosting?

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