JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.tree import DecisionTreeRegressor

X = np.array([[1], [2], [3], [4], [5]])
y = np.array([10, 20, 30, 40, 50])
model = DecisionTreeRegressor(random_state=0).fit(X, y)
print(model.predict(np.array([[100]]))[0])

50.0

Expecting 1000.0, as if the tree continued the line y = 10x: x = 100 passes every threshold like x = 5 and lands in the rightmost leaf, whose constant is 50.0.
Expecting 30.0, the mean of all targets: the tree returns the constant of the leaf the sample reaches, not the overall mean, and that leaf holds only x = 5.

What would LinearRegression predict for the same input, and why does it differ from the tree?
How can a tree-based model still be used to forecast a value that keeps growing over time?

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.tree import DecisionTreeRegressor

X = np.array([[1], [2], [3], [4], [5]])
y = np.array([10, 20, 30, 40, 50])
model = DecisionTreeRegressor(random_state=0).fit(X, y)
print(model.predict(np.array([[100]]))[0])

50.0

Expecting 1000.0, as if the tree continued the line y = 10x: x = 100 passes every threshold like x = 5 and lands in the rightmost leaf, whose constant is 50.0.
Expecting 30.0, the mean of all targets: the tree returns the constant of the leaf the sample reaches, not the overall mean, and that leaf holds only x = 5.

What would LinearRegression predict for the same input, and why does it differ from the tree?
How can a tree-based model still be used to forecast a value that keeps growing over time?

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