Turn One Dial, Refit — easy Machine Learning in Code problem | Incognition

Turn One Dial, Refit

10 pts · 10 coins

Machine Learning in CodeEasy

Turn One Dial, Refit

You have one pipeline and want to try several neighbour counts without rebuilding it each time.

Task: write neighbour_sweep(k_values).

Load the data with load_breast_cancer(return_X_y=True, as_frame=True).
Split it with train_test_split(X, y, test_size=0.25, random_state=0, stratify=y).
Build one pipeline, make_pipeline(StandardScaler(), KNeighborsClassifier()).
For each k in k_values, in order: set the pipeline's neighbour count to k, fit it on the training rows, and score it on the test rows.

Return the list of test accuracies, each rounded to 3 decimal places, in the same order as k_values.

Examples

k_values = [5]

returns [0.951]

k_values = [2, 7, 101]

returns [0.93, 0.958, 0.923]

Hints

solution.py

1from sklearn.datasets import load_breast_cancer
2from sklearn.model_selection import train_test_split
3from sklearn.neighbors import KNeighborsClassifier
4from sklearn.pipeline import make_pipeline
5from sklearn.preprocessing import StandardScaler
6
7
8def neighbour_sweep(k_values):
9    pass
10