KNNHyperparameters

Properties

k

integer | object≥ 1

Number of neighbors. Must be less than the number of training cases.

kernel

string | object

Kernel used to weight neighbors by distance; "rectangular" gives unweighted KNN.

one of"rectangular""triangular""epanechnikov""biweight""triweight""cos""inv""gaussian""rank""optimal"

distance

number | object> 0

Parameter of the Minkowski distance: 1 is Manhattan, 2 is Euclidean.

scale

boolean

Scale features to unit variance before computing distances.

ifw

boolean | object

Inverse Frequency Weighting in classification.

Relationships