EvalML builds, optimizes, and evaluates machine learning pipelines with automated data checks and objective functions.
Open source
24 alternativesOpenMetricLearning
OpenMetricLearning provides metric learning pipelines, models, and a model zoo.
About OpenMetricLearning
OpenMetricLearning (OML) is an open-source PyTorch-based framework hosted on GitHub, focused on metric learning and embedding-based retrieval tasks. It is built to train and validate models that produce embeddings optimized for distance-based similarity search, addressing limitations of standard classification models for retrieval use cases where distance metrics and top-N search performance are the primary evaluation criteria.
The framework offers configurable end-to-end training and validation pipelines, a model zoo of pretrained weights accessible via simple API calls, and native integration with PyTorch Lightning for streamlined training workflows, including support for distributed data parallel (DDP) setups. It includes built-in retrieval metrics for validation, visualization tools for search results, and pre-tested hyperparameters and examples for real-world benchmarks like large-scale product image retrieval. It is suited for data scientists, ML engineers, and researchers working on similarity search, image retrieval, or metric learning projects who want to avoid building custom pipelines and tooling from scratch.
Key features
- PyTorch-based framework for metric learning and retrieval tasks
- Configurable end-to-end training and validation pipelines
- Model zoo of pretrained weights with simple API access
- Native PyTorch Lightning integration for training workflows
- Built-in retrieval metrics for embedding validation
- Support for distributed data parallel (DDP) training
Best for: It suits data scientists, ML engineers, and researchers working on similarity search, image retrieval, or metric learning projects who want to avoid building custom training and validation pipelines from scratch.
Good for
- Low entry threshold with pre-built pipelines and pretrained models
- Pipeline-oriented design focused on practical real-world use cases
- PyTorch Lightning integration simplifies distributed training workflows
- Includes built-in retrieval metrics and search result visualization tools
OpenMetricLearning at a glance
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
OpenMetricLearning alternatives
Tools and sites you could use instead, ranked by how closely they match.
Detecto supports building and training computer vision models with the PyTorch machine learning framework.
Open source
24 alternatives
LightlyTrain supports pretraining, fine-tuning, and distillation for computer vision models.
Open source
24 alternatives
Prodmodel supports building, testing, deploying, and iterating on data science pipelines.
doddle-model is an in-memory machine learning library for Scala, built on Breeze with a scikit-learn-like API using typeclasses.
Open source
24 alternatives
AutoGluon provides tools for automated machine learning and model development.
Open source
24 alternativesPopular in ML Training & Optimization
Well-known listings from the same topic.
Edwardedwardlib.orgAerosolveaerosolve.airbnb.techDataturedatature.ioTogether AItogether.aiOptunaoptuna.orgCatBoostcatboost.ai
OpenMetricLearning FAQ
What is OpenMetricLearning used for?
OpenMetricLearning is a PyTorch-based framework for metric learning and embedding-based retrieval tasks, used to train and validate models optimized for similarity search.
How does OpenMetricLearning differ from PyTorch Metric Learning?
Unlike PyTorch Metric Learning, which is a set of metric learning tools, OpenMetricLearning focuses on end-to-end pipelines, pretrained model zoos, and practical real-world use case recipes.
Does OpenMetricLearning support distributed training?
Yes, it is integrated with PyTorch Lightning, which includes native support for distributed data parallel (DDP) training setups.
Are pretrained models available for OpenMetricLearning?
Yes, the framework includes a model zoo of pretrained weights that can be loaded via simple API calls, similar to torchvision's pretrained model access.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.