Ease of use:

  • Add metric learning to your application with just 2 lines of code in your training loop.
  • Mine pairs and triplets with a single function call.
  • Flexibility
  • Mix and match losses, miners, and trainers in ways that other libraries don't allow.

Blog posts:

Don't forget to tag @KevinMusgrave in your comment, otherwise they may not be notified.

Authors
Computer science PhD student studying computer vision and machine learning.
Share this project
Similar projects
Evaluation Metrics For Information Retrieval
Learn about common metrics used to evaluate performance of information retrieval systems
HuggingFace nlp library
Nlp is a lightweight and extensible library to easily share and load dataset and evaluation metrics, already providing access to ~100 datasets and ~10 ...
Small differences in BLEU are meaningless
Only big differences in metric scores are meaningful in MT.
BLEURT: Learning Robust Metrics for Text Generation
A metric for Natural Language Generation based on transfer learning.