Metric Learning Python Packages

pytorch-metric-learning

The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.

catalyst

Accelerated deep learning R&D

metric-learn

Metric learning algorithms in Python

open-metric-learning

Metric learning and retrieval pipelines, models and zoo.

finetuner

:dart: Task-oriented embedding tuning for BERT, CLIP, etc.

tensorflow-similarity

TensorFlow Similarity is a python package focused on making similarity learning quick and easy.

tfsim-nightly

TensorFlow Similarity is a python package focused on making similarity learning quick and easy.

disent

🧶 Modular VAE disentanglement framework for python built with PyTorch Lightning ▸ Including metrics and datasets ▸ With strongly supervised, weakly supervised and unsupervised methods ▸ Easily configured and run with Hydra config ▸ Inspired by disentanglement_lib

pylmnn

Large Margin Nearest Neighbor implementation in python

metric-learning-layers

A simple PyTorch package that includes the most common metric learning layers

powerful-benchmarker

A library for ML benchmarking. It's powerful.

ggml-ot

Improve Optimal Transport applications by Global Ground Metric Learning (published at AISTATS2025). This package supports single-cell data in Anndata format and generic array-like data.

pytorchcml

PyTorchCML is a library of PyTorch implementations of matrix factorization (MF) and collaborative metric learning (CML), algorithms used in recommendation systems and data mining.

quaterion

Similarity Learning fine-tuning framework

chesspos

A library for manipulating, learning and searching chess positions

tomotwin-cryoet

cryo-ET particle picking by representation and metric learning

gedml

Generalized Deep Metric Learning.

tf-metric-learning

Minimalistic TensorFlow2+ deep metric/similarity learning library with loss functions, miners, and utils as embedding projector.

kernelmethods

Foundational library for Kernel methods in pattern analysis and machine learning

scmdml

A multitask deep metric learning-based framework for single cell assignment based on multiple references

finetuner-client

:dart: Task-oriented embedding tuning for BERT, CLIP, etc.

ssrlib

Framework for unsupervised and self-supervised learning & embeddings analysis

catalyst-pdm

Catalyst fork compatible with PDM

wsingular

Python package for the ICML 2022 paper "Unsupervised Ground Metric Learning Using Wasserstein Singular Vectors".