dataeval.shift.OODKNeighbors¶

class dataeval.shift.OODKNeighbors(k=None, distance_metric=None, config=None)¶

K-Nearest Neighbors Out-of-Distribution detector.

Uses average distance to k nearest neighbors in embedding space to detect OOD samples. Samples with larger average distances to their k nearest neighbors in the reference (in-distribution) set are considered more likely to be OOD.

Based on the methodology from: “Back to the Basics: Revisiting Out-of-Distribution Detection Baselines” (Kuan & Mueller, 2022)

As referenced in: “Safe AI for coral reefs: Benchmarking out-of-distribution detection algorithms for coral reef image surveys”

Parameters:¶

k : int, default 10¶

Number of nearest neighbors to consider

distance_metric : "cosine" | "euclidean", default "cosine"¶

Distance metric to use

config : OODKNeighbors.Config or None, default None¶

Optional configuration object with default parameters. Parameters specified directly in __init__ will override config defaults.

Examples

>>> from dataeval.shift import OODKNeighbors
>>> import numpy as np
>>>
>>> # Create reference embeddings (in-distribution)
>>> ref_embeddings = np.random.randn(100, 128).astype(np.float32)
>>>
>>> # Fit the detector
>>> detector = OODKNeighbors(k=10, distance_metric="cosine")
>>> detector.fit(ref_embeddings, threshold_perc=95.0)
>>>
>>> # Score new samples
>>> test_embeddings = np.random.randn(20, 128).astype(np.float32)
>>> scores = detector.score(test_embeddings)
>>> predictions = detector.predict(test_embeddings)

Using configuration:

>>> config = OODKNeighbors.Config(k=15, distance_metric="euclidean", threshold_perc=99.0)
>>> detector = OODKNeighbors(config=config)
>>> detector.fit(ref_embeddings)  # Uses config.threshold_perc

fit(embeddings, threshold_perc=None)¶

Fit the detector using reference (in-distribution) embeddings.

Builds a k-NN index for efficient nearest neighbor search and computes reference scores for automatic thresholding.

Parameters:¶

embeddings : Array¶

Reference (in-distribution) embeddings

threshold_perc : float or None, default None¶

Percentage of reference data considered normal (0-100). Higher values result in more permissive thresholds. If None, uses config.threshold_perc (default 95.0).

predict(x, batch_size=int(10000000000.0), ood_type='instance')¶

Predict whether instances are out of distribution or not.

Parameters:¶

x : ArrayLike¶

Input embeddings for out-of-distribution prediction.

batch_size : int, default 1e10¶

Batch size parameter for API consistency. Not used by this detector.

ood_type : "feature" | "instance", default "instance"¶

OOD type parameter for API consistency. This detector only supports “instance” level.

Returns:¶

Dictionary containing: - is_ood: Boolean array indicating which samples are OOD - instance_score: OOD scores for all samples - feature_score: None (not supported by this detector)

Return type:¶

OODOutput

score(x, batch_size=int(10000000000.0))¶

Compute the out of distribution scores for a given dataset.

Parameters:¶

x : ArrayLike¶

Input embeddings to score.

batch_size : int, default 1e10¶

Batch size parameter for API consistency. Not used by this detector.

Returns:¶

An object containing the instance-level OOD scores. Higher scores indicate samples are more likely to be OOD.

Return type:¶

OODScoreOutput

Classes¶