dashstander/block-recurrent-transformer

Pytorch implementation of "Block Recurrent Transformers" (Hutchins & Schlag et al., 2022)

41
/ 100
Emerging

This is a tool for machine learning researchers and practitioners who are experimenting with advanced neural network architectures. It helps you build and train models that can process sequences of information more efficiently than traditional methods, particularly when dealing with very long inputs. You provide your training data, and it helps you produce a trained recurrent transformer model.

No commits in the last 6 months.

Use this if you are developing or researching new deep learning models for sequence processing and want to explore the Block Recurrent Transformer architecture.

Not ideal if you need an out-of-the-box solution for a specific application without delving into model architecture or training details.

deep-learning-research neural-networks sequence-modeling pytorch-development model-architecture
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

85

Forks

13

Language

Python

License

MIT

Last pushed

May 14, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/dashstander/block-recurrent-transformer"

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