Skip to content

KNNClassifier

K-Nearest Neighbors (KNN) for classification.

Samples are stored using a first-in, first-out strategy. The strategy to perform search queries in the data buffer is defined by the engine parameter.

Parameters

  • n_neighbors

    Type → int

    Default → 5

    The number of nearest neighbors to search for.

  • engine

    Type → BaseNN | None

    Default → None

    The search engine used to store the instances and perform search queries. Depending on the choose engine, search will be exact or approximate. Please, consult the documentation of each available search engine for more details on its usage. By default, use the SWINN search engine for approximate search queries.

  • weighted

    Type → bool

    Default → True

    Weight the contribution of each neighbor by it's inverse distance.

  • cleanup_every

    Type → int

    Default → 0

    This determines at which rate old classes are cleaned up. Classes that have been seen in the past but that are not present in the current window are dropped. Classes are never dropped when this is set to 0.

  • softmax

    Type → bool

    Default → False

    Whether or not to use softmax normalization to normalize the neighbors contributions. Votes are divided by the total number of votes if this is False.

Examples

import functools
from river import datasets
from river import evaluate
from river import metrics
from river import neighbors
from river import preprocessing
from river import utils

dataset = datasets.Phishing()

To select a custom distance metric which takes one or several parameter, you can wrap your chosen distance using functools.partial:

l1_dist = functools.partial(utils.math.minkowski_distance, p=1)

model = (
    preprocessing.StandardScaler() |
    neighbors.KNNClassifier(
        engine=neighbors.SWINN(
            dist_func=l1_dist,
            seed=42
        )
    )
)

evaluate.progressive_val_score(dataset, model, metrics.Accuracy())

Accuracy: 89.75%

Methods

clean_up_classes

Clean up classes added to the window.

Classes that are added (and removed) from the window may no longer be valid. This method cleans up the window and and ensures only known classes are added, and we do not consider "None" a class. It is called every cleanup_every step, or can be called manually.

learn_one

Update the model with a set of features x and a label y.

Parameters

  • x — 'dict'
  • y — 'base.typing.ClfTarget'

Returns

Classifier: self

predict_one

Predict the label of a set of features x.

Parameters

  • x — 'dict'
  • kwargs

Returns

base.typing.ClfTarget | None: The predicted label.

predict_proba_one

Predict the probability of each label for a dictionary of features x.

Parameters

  • x — 'dict'
  • kwargs

Returns

dict[base.typing.ClfTarget, float]: A dictionary that associates a probability which each label.

Notes

Note that since the window is moving and we keep track of all classes that are added at some point, a class might be returned in a result (with a value of 0) if it is no longer in the window. You can call model.clean_up_classes(), or set cleanup_every to a non-zero value.