pyod

library3.6.5pypypi✓ verified 25d ago

PyOD is a comprehensive and scalable Python library for outlier detection (anomaly detection), offering over 50 detection models. It provides a unified API, making it easy to use and compare various algorithms. The library is currently at version 2.1.0, with frequent minor releases addressing compatibility and adding new features, including recent advancements in multi-modal anomaly detection using foundation model embeddings.

ai-ml data

pip install pyod

github ↗package ↗docs ↗changelog ↗

INSTALL

IMPORT

SIG · PYOD

pyod

ai-mlpythonv3.6.5

Install

17.1s avg

Import

4555ms

Disk

567MB

Pass rate

5/ 10

Env Coverage5 / 10

glibc

3.9–3.13

musl

3.9–3.13

Install & Compatibility

Where this runs

tested against v3.6.5 · pip install

no network on importno background threads

Install × environment matrix

Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux

musl

py 3.10–3.910 runs

build_error

glibc

py 3.10–3.910 runs

installs and imports cleanly · install 17.1s · import 4.555s · 556MB

567MB installed

● package 567MB

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

KNN

✓ from pyod.models.knn import KNN

generate_data

✓ from pyod.utils.data import generate_data

EmbeddingOD

✓ from pyod.models.embedding_od import EmbeddingOD

New in v2.1.0 for multi-modal anomaly detection, requires additional dependencies like `sentence-transformers`.

This quickstart demonstrates how to generate synthetic data and use the k-Nearest Neighbors (KNN) algorithm from PyOD to detect outliers. It shows the basic steps of initialization, fitting the model, and retrieving binary outlier labels and raw anomaly scores.

from pyod.models.knn import KNN from pyod.utils.data import generate_data import numpy as np # Generate random data with 20% outliers X_train, y_train = generate_data(n_train=200, n_features=2, n_outliers=20, random_state=42) # Initialize and train a kNN detector clf = KNN(contamination=0.1) # Set contamination based on expected outlier ratio clf.fit(X_train) # Get the prediction labels (0: inliers, 1: outliers) y_train_pred = clf.labels_ # Get the raw outlier scores y_train_scores = clf.decision_scores_ print(f"Number of training samples: {len(X_train)}") print(f"Number of predicted outliers: {np.count_nonzero(y_train_pred)}")

Debug

Known issues

breakingPyOD removed all TensorFlow and Keras code, migrating deep learning models entirely to PyTorch.

fix

If you relied on TensorFlow-based models, either pin your PyOD version to <2.0.2 or refactor your code to use PyTorch-based models available in PyOD v2.0.2 and later.

affects: <2.0.2

breakingDefault parameters for some models, notably VAE, have changed (e.g., output activation for VAE to `identity`).

fix

Existing VAE models trained with older PyOD versions may produce different results. Consider re-training models or explicitly setting parameters like `output_activation` to maintain backward compatibility if migrating.

affects: <2.0.7

gotchaThe `contamination` parameter is crucial and can significantly impact detection results and thresholds, especially when not set accurately.

fix

Understand that `contamination` is the proportion of outliers in the data. It's used for thresholding (`predict` and `labels_`). If unknown, careful validation or methods like `predict_proba` might be needed. Setting it too high or too low can lead to misclassifications.

affects: All versions

gotchaPyOD frequently updates its internal dependencies and sometimes makes adjustments for `scikit-learn` breaking changes.

fix

It's generally recommended to keep PyOD updated, or ensure compatibility between your `pyod` and `scikit-learn` versions to avoid unexpected behavior, especially after major `scikit-learn` releases.

affects: All versions, particularly pre-2.0.6 with newer scikit-learn

gotchaThe new `EmbeddingOD` framework (v2.1.0+) requires additional, potentially large, third-party libraries (e.g., `sentence-transformers`, `openai`, `transformers`) that are not installed by default.

fix

Install the necessary optional dependencies using `pip install pyod[text]` or `pip install pyod[image]` or individual packages as required by your chosen embedding model (e.g., `pip install sentence-transformers`).

affects: 2.1.0+

Errors

Common errors & fixes

ModuleNotFoundError: No module named 'pyod'

The PyOD library is not installed in the current Python environment.

fix

pip install pyod

ImportError: cannot import name 'LOF' from 'pyod.models.lof'

PyOD model classes (e.g., LOF) are typically exposed directly under the 'pyod.models' module, not nested within submodules.

fix

from pyod.models import LOF

ValueError: Expected 2D array, got 1D array instead:

PyOD models, like most scikit-learn estimators, expect input data (X) to be a 2D array (n_samples, n_features), but a 1D array was provided.

fix

Reshape your input data to a 2D array, for example, using 'X.reshape(-1, 1)' for a single feature.

AttributeError: 'LOF' object has no attribute 'decision_function'

You are attempting to access a non-existent or incorrectly named attribute to retrieve anomaly scores after fitting a PyOD model.

fix

Use 'model.decision_scores_' to get raw anomaly scores or 'model.labels_' for binary outlier labels after fitting.

ValueError: Expected 2D array, got 1D array instead: array=[...]. Reshape your data using X.reshape(-1, 1) if it contains a single feature or X.reshape(1, -1) if it contains a single sample.

PyOD models, like scikit-learn estimators, expect input data to be a 2D array (n_samples, n_features), but a 1D array (e.g., a single feature or a single sample) was provided.

fix

Reshape your 1D input data `X` into a 2D array, for example, `X_train_reshaped = X_train.reshape(-1, 1)` if it's a feature array, or convert a pandas Series using `X_train_reshaped = X_train.values.reshape(-1, 1)`.

Upgrade

Version history

3.6.5latest on PyPI · released Aug 17, 2026

Audit

Dependencies

scikit-learnrequiredCore machine learning utilities and base estimators.

torchrequiredRequired for deep learning-based models, replacing TensorFlow since v2.0.2.

sentence-transformersoptionalOptional, required for text embeddings in EmbeddingOD (v2.1.0+).

transformersoptionalOptional, for HuggingFace model embeddings in EmbeddingOD (v2.1.0+).

openaioptionalOptional, for OpenAI model embeddings in EmbeddingOD (v2.1.0+).

Agent activity

32 hits · last 30 days

node

OpenAI (training)

Resources

githubhttps://github.com/yzhao062/pyod ↗

packagehttps://pypi.org/project/pyod/ ↗

docshttps://pyod.readthedocs.io ↗

changeloghttps://github.com/yzhao062/pyod/blob/development/CHANGES.txt ↗