awesome-repositories.com
Blog
awesome-repositories.com

Découvrez les meilleurs dépôts open-source grâce à notre recherche par IA.

ExplorerRecherches sélectionnéesAlternatives open sourceLogiciels auto-hébergésBlogPlan du site
ProjetÀ proposNotre méthodologiePresseServeur MCP
Mentions légalesConfidentialitéConditions d'utilisation
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·

13 dépôts

Awesome GitHub RepositoriesAnomaly Detection Algorithms

Algorithms for identifying outliers or unusual patterns in datasets.

Distinct from Anomaly Detection: Focuses on algorithmic detection of anomalies in sequences rather than statistical or log-based monitoring.

Explore 13 awesome GitHub repositories matching data & databases · Anomaly Detection Algorithms. Refine with filters or upvote what's useful.

Awesome Anomaly Detection Algorithms GitHub Repositories

Trouvez les meilleurs dépôts grâce à l'IA.Nous recherchons les dépôts les plus pertinents grâce à l'IA.
  • microsoft/bringing-old-photos-back-to-lifeAvatar de microsoft

    microsoft/Bringing-Old-Photos-Back-to-Life

    15,691Voir sur GitHub↗

    This project is a deep learning image restoration tool designed to remove scratches, fading, and noise from aged photographs and film. It utilizes generative adversarial networks for image translation, alongside specialized networks for face enhancement and video colorization. The system distinguishes itself through a combination of latent-space domain mapping and progressive face enhancement to recover blurred or missing high-frequency facial details. For video content, it employs a colorization framework that uses optical flow and temporal guidance to propagate color from selected keyframes

    Identifies image defects by comparing current frame features against learned representations of neighboring frames without manual labels.

    Pythongansgenerative-adversarial-networkimage-manipulation
    Voir sur GitHub↗15,691
  • mission-peace/interviewAvatar de mission-peace

    mission-peace/interview

    11,306Voir sur GitHub↗

    This project is a comprehensive library of reference implementations for fundamental data structures and algorithms, designed to support technical interview preparation and software engineering assessments. It provides a structured collection of computational techniques for solving complex problems involving arrays, strings, graphs, trees, and mathematical analysis. The library distinguishes itself by offering specialized implementations for advanced topics, including concurrent programming patterns and geometric algorithms. It features thread-safe primitives for managing shared state and tas

    Implements algorithms for detecting missing values and numerical anomalies within ordered sequences.

    Java
    Voir sur GitHub↗11,306
  • yzhao062/pyodAvatar de yzhao062

    yzhao062/pyod

    9,878Voir sur GitHub↗

    PyOD is a Python anomaly detection library used to identify outliers in tabular, time series, graph, text, and image data. It provides a collection of algorithms for detecting anomalous data points and includes a unified detector interface that standardizes input and output signatures across its available detection algorithms. The project features a multi-modal outlier detector for identifying anomalies across diverse formats including unstructured text and images, as well as a specialized toolkit for graph-based and time-series anomaly detection. It includes an ensemble framework for combini

    Implements anomaly detection using Variational Autoencoder architectures to identify outliers via reconstruction error.

    Pythonagentic-aianomaly-detectiondata-mining
    Voir sur GitHub↗9,878
  • lawlite19/machinelearning_pythonAvatar de lawlite19

    lawlite19/MachineLearning_Python

    8,526Voir sur GitHub↗

    This is a Python machine learning library featuring a collection of core algorithms implemented from scratch to demonstrate foundational AI concepts. It provides a comprehensive toolkit for supervised learning, unsupervised learning, and neural network development. The project is distinguished by its custom implementation of a neural network framework, which includes multi-layer perceptrons with backpropagation, gradient descent, and weight regularization. It also includes a specialized anomaly detection toolkit that identifies outliers and rare events using Gaussian probability distributions

    Ships a specialized anomaly detection toolkit using Gaussian distributions and F1-score optimization.

    Python
    Voir sur GitHub↗8,526
  • priorlabs/tabpfnAvatar de PriorLabs

    PriorLabs/TabPFN

    7,408Voir sur GitHub↗

    Identifies outlier rows in tabular datasets by flagging samples with low probability under a learned distribution.

    Pythondata-sciencefoundation-modelsmachine-learning
    Voir sur GitHub↗7,408
  • greyhatguy007/machine-learning-specialization-courseraAvatar de greyhatguy007

    greyhatguy007/Machine-Learning-Specialization-Coursera

    6,996Voir sur GitHub↗

    This repository is a collection of implementation references and solved notebooks covering supervised, unsupervised, and reinforcement learning techniques. It provides practical guides for building predictive models, clustering algorithms, and autonomous agents. The project includes specific implementations for neural network architectures, such as multi-layer perceptrons for digit recognition, and recommender systems using collaborative and content-based filtering. It also features reinforcement learning systems that utilize deep Q-learning to optimize decision-making policies. The codebase

    Implements Gaussian anomaly detection to identify outliers by modeling the distribution of normal data points.

    Jupyter Notebookandrew-ngandrew-ng-machine-learningcoursera
    Voir sur GitHub↗6,996
  • numenta/nupicAvatar de numenta

    numenta/nupic

    6,352Voir sur GitHub↗

    NuPIC is a machine learning framework that implements Hierarchical Temporal Memory (HTM) theory, a neuroscience-inspired approach to artificial intelligence. It models principles of the neocortex to build systems capable of learning patterns from streaming data, performing sequence prediction, and detecting anomalies in real-time data streams. The framework is built around a Cortical Learning Algorithm that combines spatial pooling and temporal memory to process streaming input. It uses Sparse Distributed Representations to encode input patterns, a Spatial Pooler to convert dense input into s

    Flags unusual patterns in real-time data streams using neocortex-inspired temporal memory algorithms.

    Python
    Voir sur GitHub↗6,352
  • facebookresearch/katsAvatar de facebookresearch

    facebookresearch/Kats

    6,311Voir sur GitHub↗

    Kats est un framework et une bibliothèque d'analyse de séries temporelles fournissant des outils pour la caractérisation statistique, la détection d'anomalies et la prévision de tendances. Il fonctionne comme un toolkit pour prédire des valeurs futures basées sur des données historiques et identifier des modèles irréguliers ou des points de changement structurel au sein de séquences temporelles. Le projet inclut un outil d'extraction de caractéristiques temporelles pour calculer des statistiques descriptives et des caractéristiques qui résument le comportement des séries temporelles. Il fournit également un système pour le réglage des hyperparamètres de modèle utilisant l'apprentissage auto-supervisé pour améliorer l'échelle et la généralisation des prédictions.

    Provides a set of algorithms for identifying outliers and structural change points within temporal data sequences.

    Python
    Voir sur GitHub↗6,311
  • open-edge-platform/anomalibAvatar de open-edge-platform

    open-edge-platform/anomalib

    5,871Voir sur GitHub↗

    Anomalib is a PyTorch-based library for visual anomaly detection, offering a modular framework, a comprehensive model zoo, and a benchmarking suite designed for industrial defect detection. It provides a wide range of algorithms—including generative, discriminative, teacher-student, and vision-language approaches—that support unsupervised, few-shot, and zero-shot settings. The library enables deployment through model export to ONNX and OpenVINO for edge devices, and includes a no-code web application for training and inference. It also features a command-line interface for orchestrating multi

    Provides a benchmarking suite that evaluates multiple anomaly detection models across standard datasets.

    Pythonanomaly-detectionanomaly-localizationanomaly-segmentation
    Voir sur GitHub↗5,871
  • apachecn/sklearn-doc-zhAvatar de apachecn

    apachecn/sklearn-doc-zh

    5,231Voir sur GitHub↗

    Ce projet fournit une version traduite des guides et références API de la bibliothèque de machine learning scikit-learn pour les sinophones. Il sert de base de connaissances localisée et de référence technique pour implémenter l'analyse prédictive de données et la modélisation statistique en utilisant une boîte à outils basée sur Python. La ressource couvre l'implémentation de l'apprentissage supervisé, incluant les tâches de classification et de régression, ainsi que les workflows d'apprentissage non supervisé pour la découverte de motifs et la détection d'anomalies. Elle fournit également des conseils sur l'éducation en data science, en se concentrant spécifiquement sur l'utilisation de scikit-learn pour le machine learning. La documentation inclut des instructions détaillées sur le prétraitement des données, la réduction de dimensionnalité et la sélection de caractéristiques. Elle détaille en outre l'évaluation et le réglage des modèles via des métriques de performance, l'optimisation des hyperparamètres et la validation de généralisation, ainsi que l'utilisation de pipelines de prédiction et d'utilitaires de traitement du langage naturel.

    Provides instructions on identifying unusual data points using algorithms such as isolation forests and local outlier factors.

    CSSdocumentationmachine-learningpython
    Voir sur GitHub↗5,231
  • microsoft/synapsemlAvatar de microsoft

    microsoft/SynapseML

    5,230Voir sur GitHub↗

    SynapseML est une bibliothèque de machine learning Apache Spark conçue pour construire et mettre à l'échelle des workflows de machine learning et des pipelines de données à travers des clusters distribués. Elle sert de framework de pipeline de machine learning distribué et de moteur d'inférence distribué pour exécuter des prédictions accélérées par le matériel et des tâches de deep learning sur des jeux de données à grande échelle. Le projet fonctionne comme une couche d'intégration d'IA cloud, permettant aux utilisateurs d'appliquer des services d'intelligence artificielle pré-entraînés pour le texte, la vision et la parole au sein de pipelines distribués. Il inclut également une suite dédiée d'outils pour la détection distribuée d'anomalies afin d'identifier les valeurs aberrantes multivariées et de séries temporelles à travers des données de haute dimension. La bibliothèque couvre un large éventail de capacités, incluant la vision par ordinateur distribuée pour l'analyse de visages et d'images, le traitement du langage naturel évolutif pour l'analyse et la traduction de texte, et l'entraînement d'arbres de décision à gradient boosté. Elle fournit des outils pour la recherche de similarité via la modélisation k-plus proches voisins, l'explicabilité des modèles via l'attribution de caractéristiques, et l'orchestration de workflows d'apprentissage par renforcement. Le système utilise une architecture de pipeline composable et prend en charge l'inférence de modèle basée sur ONNX pour une compatibilité multiplateforme.

    Implements algorithms to identify outliers across multiple data streams by analyzing inter-correlations and dependencies.

    Scalaaiapache-sparkazure
    Voir sur GitHub↗5,230
  • okfn-brasil/serenata-de-amorAvatar de okfn-brasil

    okfn-brasil/serenata-de-amor

    4,595Voir sur GitHub↗

    Ce projet est une plateforme d'audit des dépenses publiques et de contrôle social pilotée par l'IA, conçue pour surveiller l'administration publique et signaler les activités financières suspectes. Il fonctionne comme un outil automatisé de détection de la fraude qui utilise la reconnaissance de formes pour identifier les irrégularités dans les registres de dépenses gouvernementales. Le système transforme les fichiers de dépenses publiques bruts en jeux de données structurés via un pipeline de données, mappant divers formats de données gouvernementales vers un modèle relationnel standardisé. Il utilise un système de signalement automatisé pour prioriser l'examen manuel par les auditeurs et génère des exportations de jeux de données versionnées basées sur des instantanés pour garantir la reproductibilité à des fins de vérification indépendante. La plateforme inclut une interface de visualisation de données destinée au public, permettant aux citoyens d'explorer et de valider des registres de dépenses détaillés. Cette capacité soutient la responsabilité du secteur public en exposant les modèles identifiés lors de l'analyse automatisée.

    Implements algorithms to identify outliers and suspicious financial patterns in government spending datasets.

    Pythonartificial-intelligencecivic-techdata-science
    Voir sur GitHub↗4,595
  • hosseinmoein/dataframeAvatar de hosseinmoein

    hosseinmoein/DataFrame

    2,917Voir sur GitHub↗

    DataFrame is a C++ tabular data library and manipulation engine designed for managing heterogeneous data in contiguous memory. It functions as a statistical analysis framework and time series analysis toolkit, providing the means to store, index, and transform multidimensional datasets. The project distinguishes itself through a high-performance execution model that utilizes column-major storage, SIMD-aligned memory allocation, and a thread-pool for parallel computations. It employs a visitor-based algorithm dispatch system and policy-driven transformations to decouple data processing logic f

    Flags outlier values that exceed a defined number of standard deviations from the mean.

    C++aicppdata-analysis
    Voir sur GitHub↗2,917
  1. Home
  2. Data & Databases
  3. Anomaly Detection Algorithms

Explorer les sous-tags

  • Adversarial Anomaly DetectorsConditional GAN-based models that reconstruct normal images and detect anomalies by large reconstruction errors. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: focuses specifically on adversarial (GAN) approaches for visual anomaly detection rather than general algorithmic methods.
  • Anomaly Detection Dataset LoadersLoads and prepares visual anomaly detection data from standard benchmarks and custom folders, supporting multiple modalities. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: focuses on data loading and preparation, not the detection algorithms themselves.
  • Benchmarking SuitesTools for evaluating multiple anomaly detection models across standard datasets with comprehensive metrics. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: focuses on cross-model evaluation and comparison, not the algorithms themselves.
  • Benchmarking ToolsTools for evaluating state-of-the-art anomaly detection methods on public and private datasets. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: focuses on the evaluation and comparison of algorithms, not the algorithms themselves.
  • Coupled Hypersphere DetectorsAdapts image features to a coupled hypersphere target space to detect and localize visual anomalies. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses a coupled hypersphere representation for joint detection and localization.
  • Custom Model FrameworksCore building blocks and utilities for assembling custom anomaly detection models from scratch. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: provides the framework for building custom models rather than pre-built algorithm implementations.
  • Deep Feature Anomaly DetectorsExtracts deep features using a pre-trained CNN backbone and applies PCA with a Gaussian model for detection. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses deep feature extraction with PCA and Gaussian modeling.
  • Deep Feature Density EstimatorsExtracts deep features from a pretrained backbone and fits a kernel density model on normal samples to score anomalies. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically combines deep feature extraction with kernel density estimation for scoring.
  • Discriminative Anomaly DetectorsTrains a discriminative model to produce pixel-level anomaly maps and global scores for surface defect detection. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: uses a discriminative approach to produce both pixel-level and image-level anomaly predictions.
  • Fast Anomaly DetectorsUses a student-teacher architecture with a pre-trained EfficientNet backbone to achieve millisecond-speed inference. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically optimized for real-time inference speed using a student-teacher design.
  • Feature Distillation DetectorsUses a Vision Transformer teacher and two student MLPs to detect anomalies via reconstruction error from feature distillation. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses knowledge distillation between ViT and MLP students.
  • Feature Pyramid Matching DetectorsTrains a student network to match multi-scale features from a pretrained teacher and flags discrepancies as anomalies. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses multi-scale feature pyramid matching between teacher and student.
  • Feature Reconstruction DetectorsDetects anomalous regions by measuring reconstruction error between original and reconstructed features through a tied autoencoder. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses tied autoencoder for feature reconstruction error.
  • Few-Shot Anomaly Detectors1 sous-tagPerforms anomaly classification and segmentation by comparing image regions to normal reference images using CLIP embeddings with few examples. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically supports few-shot and zero-shot scenarios using CLIP embeddings.
  • Gaussian Anomaly DetectorsStatistical methods that use Gaussian distributions to identify outliers based on probability density. **Distinct from Anomaly Detection Algorithms:** Specializes anomaly detection by specifically using Gaussian distribution modeling rather than general algorithmic patterns.
  • Normal Prototype DetectorsLearns per-image normal prototypes from training on normal images and uses reconstruction error to localize anomalies. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically learns prototypes of normal images for reconstruction-based anomaly localization.
  • Normalizing Flow Anomaly DetectorsAnomaly detection algorithms that model normal data distributions using normalizing flows for efficient inference. **Distinct from Anomaly Detection Algorithms:** Distinct from general Anomaly Detection Algorithms: specifically uses normalizing flows to model distributions, enabling real-time detection.
  • Normalizing Flow Detectors1 sous-tagLearns the distribution of normal patch features with a normalizing flow to produce image-level and pixel-wise anomaly detections. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses normalizing flows to model the distribution of normal features.
  • Patch Discrimination DetectorsTrains a patch-level discriminator on frozen backbone features to score image regions and aggregate them for anomaly detection. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically uses a patch-level discriminator on frozen backbone features.
  • Patch Distribution DetectorsFits per-location Gaussian distributions to patch embeddings from a pretrained CNN to detect and localize anomalies. **Distinct from Anomaly Detection Algorithms:** Distinct from general Anomaly Detection Algorithms: specifically uses per-patch Gaussian modeling on CNN features.
  • Patch k-Nearest Neighbors DetectorsDetects anomalies by comparing patch embeddings to a memory bank of normal patches using k-nearest neighbor search. **Distinct from Anomaly Detection Algorithms:** Distinct from general Anomaly Detection Algorithms: uses patch-level kNN search against a normal memory bank.
  • Quantized Feature Anomaly DetectorsAlgorithms that learn a quantized feature codebook and generate synthetic defects to train decoders for anomaly detection without external defect datasets. **Distinct from Deep Feature Anomaly Detectors:** Distinct from Deep Feature Anomaly Detectors: uses a quantized codebook and synthetic defect generation rather than PCA and Gaussian modeling on deep features.
  • RGB-D Anomaly DetectorsAlgorithms that jointly process color and depth data to detect visual anomalies. **Distinct from Anomaly Detection Algorithms:** Distinct from general anomaly detection algorithms: specifically processes RGB-D data (color + depth) for visual anomaly detection.
  • Reconstruction-Based Anomaly DetectorsAnomaly detection algorithms that compare input images with their reconstructions to identify anomalous regions. **Distinct from Feature Reconstruction Detectors:** Distinct from Feature Reconstruction Detectors: uses a discriminatively trained embedding network for reconstruction, not a tied autoencoder.
  • Reverse Distillation Anomaly DetectorsAnomaly detection algorithms that use a decoder to reconstruct multi-scale features from a compressed bottleneck and score anomalies by reconstruction error. **Distinct from Anomaly Detection Algorithms:** Distinct from general Anomaly Detection Algorithms: specifically uses a reverse distillation architecture with a compressed bottleneck for feature reconstruction.
  • Severity ScoringQuantifying the degree of anomaly for a data point relative to the dataset distribution. **Distinct from Anomaly Detection Algorithms:** Focuses on the severity/degree of an anomaly rather than a binary label or general algorithm.
  • Streaming Anomaly Detection Engines1 sous-tagEngines that flag unusual patterns in real-time data streams using neocortex-inspired temporal memory algorithms. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: focuses on streaming anomaly detection engines, not general algorithmic detection.
  • Structural Graph Anomaly DetectionDetecting anomalies in graphs by analyzing structural connectivity and node attributes. **Distinct from Anomaly Detection Algorithms:** Specifically targets structural anomalies using dual autoencoders in graph data
  • Tabular Anomaly DetectorsIdentifies outlier rows in tabular datasets by flagging samples with low probability under a learned distribution. **Distinct from Anomaly Detection Algorithms:** Distinct from Anomaly Detection Algorithms: specifically targets tabular data using learned probability distributions.
  • ToolkitsIntegrated sets of algorithms and utilities for identifying anomalies and outliers. **Distinct from Anomaly Detection Algorithms:** Distinct from individual algorithms by providing a cohesive toolkit including scoring and optimization
  • Unsupervised Anomaly Detectors1 sous-tagAlgorithms that detect anomalies without requiring labeled training samples, often by synthesizing defects. **Distinct from Anomaly Detection Algorithms:** Distinct from general anomaly detection algorithms: specifically operates without labeled anomalies by synthesizing defects during training.
  • Variational AutoencodersOutlier detection using variational autoencoder architectures to flag high reconstruction errors. **Distinct from Denoising Auto-Encoders:** Specific VAE implementation for anomaly detection, distinct from general denoising auto-encoders.