A Score With an Error Bar — easy Machine Learning in Code problem | Incognition

A Score With an Error Bar

10 pts · 10 coins

Machine Learning in CodeEasy

A Score With an Error Bar

One train/test split gives one opinion. Cross-validation gives several, and their spread tells you how much to trust the average.

Task: write cv_mean_and_spread(n_folds).

Load the data with load_breast_cancer(return_X_y=True, as_frame=True).
Build make_pipeline(StandardScaler(), LogisticRegression(max_iter=1000)).
Score it with cross_val_score on all the rows, using cv=n_folds.

Return [mean, spread]: the mean of the fold scores and their standard deviation (NumPy's .std()), each rounded to 3 decimal places.

Examples

n_folds = 5

returns [0.981, 0.007]

n_folds = 8

returns [0.981, 0.012]

Hints

solution.py

1from sklearn.datasets import load_breast_cancer
2from sklearn.linear_model import LogisticRegression
3from sklearn.model_selection import cross_val_score
4from sklearn.pipeline import make_pipeline
5from sklearn.preprocessing import StandardScaler
6
7
8def cv_mean_and_spread(n_folds):
9    pass
10