anomalib

We have hosted the application anomalib in order to run this application in our online workstations with Wine or directly.

Quick description about anomalib:

Anomalib is an open-source deep learning library focused on anomaly detection and localization tasks, collecting state-of-the-art algorithms and tools under one modular framework. It provides implementations of leading anomaly detection methods drawn from current research, as well as a full set of utilities for training, evaluating, benchmarking, and deploying these models on both public and private datasets. Anomalib emphasizes flexibility and reproducibility: you can use its simple APIs to plug in custom models, track experiments, tune hyperparameters, and generate visualizations that highlight anomalous regions. Its design supports unsupervised or semi-supervised paradigms, making it especially powerful for scenarios where only �normal� data is readily available and defects must be detected without exhaustive labeling. Combined with its CLI and integration with optimization tools like OpenVINO, it�s suitable for both research and edge deployment tasks.

Features:

• Collection of state-of-the-art anomaly detection models
• Modular API for training, inference, benchmarking
• Hyperparameter optimization and experiment tracking
• Visualization tools for anomaly localization
• CLI support for common workflows
• Export models for accelerated inference on OpenVINO

Programming Language: Python.
Categories: