Scale, Then Vote

Nearest neighbours measures distance, so it needs scaled columns. Do the scaling by hand, the honest way.

Task: write honest_score(name, k, seed). name is "iris", "wine" or "breast_cancer".

• Load the dataset as X and y, and split with train_test_split(X, y, test_size=0.25, random_state=seed).
• Fit a StandardScaler on the training rows only. Use it to scale the training rows and the test rows.
• Train a KNeighborsClassifier(n_neighbors=k) on the scaled training rows.
• Return its score on the scaled test rows, rounded to 3 decimal places.