lucidrains/PoPE-pytorch

Efficient implementation (and explorations) into polar coordinate positional embedding (PoPE) - from Gopalakrishnan et al. under Schmidhuber

55
/ 100
Established

This project helps machine learning engineers improve the performance of attention mechanisms in models that process sequences, images, or videos. It takes in sequence data or multi-dimensional data like images and outputs positionally enhanced queries and keys, which leads to better context understanding. It's for researchers and practitioners who build and optimize deep learning models.

57 stars and 1,549 monthly downloads. Used by 2 other packages. Available on PyPI.

Use this if you are a machine learning engineer working with transformer models and want to enhance how your models understand the position of elements in sequences or multi-dimensional data like images or videos.

Not ideal if you are not directly involved in developing or optimizing deep learning models that use attention mechanisms.

deep-learning-optimization transformer-architecture sequence-modeling computer-vision natural-language-processing
Maintenance 13 / 25
Adoption 17 / 25
Maturity 20 / 25
Community 5 / 25

How are scores calculated?

Stars

57

Forks

2

Language

Python

License

MIT

Last pushed

Mar 25, 2026

Monthly downloads

1,549

Commits (30d)

0

Dependencies

3

Reverse dependents

2

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/lucidrains/PoPE-pytorch"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.