k-Nearest Neighbors Lab

Upload your own CSV (or pick a sample dataset), set the kNN parameters — number of neighbours, distance metric, weighting, feature scaling — or simply keep the defaults, and classify directly in your browser.

1 · Data

Sample dataset

…or upload a CSV (last column = label)

Paste CSV instead

2 · kNN parameters

Leave everything untouched to run with the standard defaults.

n_neighbors (k)

Number of neighbours voting. Default 5.

metric

Distance function. Default euclidean.

minkowski p

Only used when metric = minkowski.

weights

Vote weighting. Default uniform.

feature scaling

kNN is scale sensitive. Default z-score.

test size

Stratified hold-out fraction. Default 0.25.

random_state

Seed for the split shuffle.

k sweep max (0 = off)

Also plot accuracy for k = 1…max.

k-Nearest Neighbors Lab

1 · Data

Class distribution

Preview (first 5 rows)

2 · kNN parameters

3 · Results

Confusion matrix (test set)

Per-class report (test set)

Permutation feature importance (accuracy drop)

Accuracy vs. k