Knn

polars_ts.classification.knn

K-Nearest Neighbors classification for time series using precomputed distances.

TimeSeriesKNNClassifier

K-Nearest Neighbors time series classifier.

Parameters

k Number of nearest neighbors. Default 3. metric Distance metric name (e.g. "dtw", "erp", "lcss"). Default "dtw". **distance_kwargs Extra keyword arguments forwarded to the distance function.

fit(df, *, label_col='label')

Fit the classifier with training data.

Parameters

df Training DataFrame with columns unique_id, y, and label_col. label_col Name of the column containing class labels.

Returns

self

predict(df)

Predict class labels for test time series.

Parameters

df Test DataFrame with columns unique_id and y.

Returns

pl.DataFrame DataFrame with columns unique_id and predicted_label.

knn_classify(train_df, test_df, k=3, method='dtw', id_col='unique_id', target_col='y', label_col='label', **distance_kwargs)

K-Nearest Neighbors time series classification.

Parameters

train_df Training DataFrame with columns id_col, target_col, and label_col. The label_col must have one label per id_col. test_df Test DataFrame with columns id_col and target_col. k Number of nearest neighbors. method Distance metric name (e.g. "dtw", "erp", "lcss"). id_col Column identifying each time series. target_col Column with the time series values. label_col Column with class labels in the training data. **distance_kwargs Extra keyword arguments forwarded to the distance function.

Returns

pl.DataFrame DataFrame with columns [id_col, "predicted_label"].