OpenMetricLearning
A PyTorch-based framework for training and validating models that produce high-quality embeddings for metric learning and retrieval tasks.
What is OpenMetricLearning?
Open Metric Learning (OML) is a PyTorch-based framework for training and validating deep learning models that produce high-quality embeddings, specifically designed for metric learning and retrieval tasks. It solves the problem of optimizing models for distance-based search scenarios, where standard classification training doesn't directly optimize retrieval metrics like cosine or L2 distances.
Machine learning engineers and researchers working on retrieval systems, person re-identification, face recognition, product search, or any application requiring similarity search over embeddings, especially those with many classes but few samples per class.
Developers choose OML for its practical, pipeline-oriented approach that includes config-based training, a zoo of pretrained models, and comprehensive tooling for metric learning, reducing the implementation overhead compared to building custom solutions or using more tool-focused libraries.
Overview
Metric learning and retrieval pipelines, models and zoo.
Use Cases
Best For
- Training models for person or animal re-identification systems
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
