Validate k on a meaningful scale

Understand the idea

Validation rows must not become training neighbours. Scaling belongs inside the candidate pipeline so each fold learns its own metric scale.

Illustrative data

Illustrative data
Column	Values
Class A	n2: (0.5, 0.2); n3: (0.4, 0.5); n6: (1.5, 1.2); n7: (-1.5, -0.2); n9: (2.9, 0.9)
Class B	n1: (0.3, 0.2); n4: (1, 0.4); n5: (-0.9, 0.5); n8: (2.4, 0)

Click the highlighted code to change the result.

example_model = Pipeline([
    ("prepare", StandardScaler()),
    ("model", KNeighborsClassifier(n_neighbors=)),
])
example_model.fit(example_X, example_y)
example_model.predict(pd.DataFrame({"x": [], "y": []}))

k=3: 2 A + 1 B → A; nearest rows n1, n2, n7. Distances use training standard deviations.

Supplied for this practice

These objects are prepared afresh before your editor runs.

X: Feature dataframe for the declared population, preserving row indices.
X = df[['length', 'width']]
y: Target series, aligned with X.
y = df.label
X_train / X_test / y_train / y_test: Training feature rows from the declared split. Final-test feature rows from the declared split. Training targets, aligned with X_train. Final-test targets, aligned with X_test.
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)

Outputs you create

answer: The requested result, with the values, labels and row order described by this task.

CLASS180 · first 8 rows
`length`	`width`	`label`
236.033	-0.805568	A
581.297	0.728558	A
-1511.27	-1.00866	A
99.0248	-0.24496	A
-13.0141	-0.660765	A
681.179	0.602475	A
51.1472	0.873157	A
362.131	-0.665605	A

Your task

Search production k values 3,5,9 on CLASS180.

answer

Hints

Think: Scaling must be learned separately inside each validation fit.

Tools: Pipeline, GridSearchCV and model__n_neighbors.

Approach: Compare the declared k values in a scaling/KNN pipeline using seeded stratified macro-F1 folds.

Explained solution

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import GridSearchCV,StratifiedKFold
model=Pipeline([('scale',StandardScaler()),('model',KNeighborsClassifier())])
search=GridSearchCV(model,{'model__n_neighbors':[3,5,9]},cv=StratifiedKFold(5,shuffle=True,random_state=42),scoring='f1_macro').fit(X_train,y_train)
answer=search.cv_results_['mean_test_score']

Fold-local scaling keeps validation observations from changing distances used to fit their own predictor.