JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.dummy import DummyClassifier

X_train = np.zeros((10, 1))
y_train = np.array([0, 0, 0, 0, 0, 0, 0, 1, 1, 1])
X_test = np.zeros((4, 1))
y_test = np.array([1, 1, 1, 0])
model = DummyClassifier()
model.fit(X_train, y_train)
print(model.score(X_test, y_test))

0.25

Expecting 0.75: the class is picked once, at fit, from y_train, where 0 is the most frequent; the test set's own majority 1 plays no part.
Expecting 0.7, the share of 0 in y_train: score compares the constant answer 0 with y_test, where only one label of four is 0.

Why train and report a baseline like this before any real model?
On a dataset with 95% negatives, what accuracy does this baseline reach, and which metric would expose it?

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.dummy import DummyClassifier

X_train = np.zeros((10, 1))
y_train = np.array([0, 0, 0, 0, 0, 0, 0, 1, 1, 1])
X_test = np.zeros((4, 1))
y_test = np.array([1, 1, 1, 0])
model = DummyClassifier()
model.fit(X_train, y_train)
print(model.score(X_test, y_test))

0.25

Expecting 0.75: the class is picked once, at fit, from y_train, where 0 is the most frequent; the test set's own majority 1 plays no part.
Expecting 0.7, the share of 0 in y_train: score compares the constant answer 0 with y_test, where only one label of four is 0.

Why train and report a baseline like this before any real model?
On a dataset with 95% negatives, what accuracy does this baseline reach, and which metric would expose it?

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