AIForce/train-dalle-together @ 748f20991cb4ecba1429963c8bbf93559ba4cfe9

Žiadny popis

7 Vetvy

Aleksandr Borzunov 748f20991c Make runners executable and improve their shebangs		3 rokov pred
lib	64dee420da Upgrade to using hivemind.optim.experimental	3 rokov pred
.gitignore	72fc0bcdb7 Initial commit (ru-max branch without private code)	3 rokov pred
README.md	d4140807f4 Update readme	3 rokov pred
arguments.py	19e3d2d060 Make aux peers fetch checkpoints every 2 steps by default	3 rokov pred
callback.py	4918c58cb6 Polish stdout	3 rokov pred
data.py	c61c61b20d Use t5-small tokenizer	3 rokov pred
huggingface_auth.py	3e604bc1f5 fix auth	3 rokov pred
manage_scaleset.py	c365b2ec9f Tweak settings for the upcoming demo (#2)	3 rokov pred
requirements.txt	3db71b9de5 Fix transformers version	3 rokov pred
run_aux_peer.py	748f20991c Make runners executable and improve their shebangs	3 rokov pred
run_trainer.py	748f20991c Make runners executable and improve their shebangs	3 rokov pred
run_trainer_tpu.py	748f20991c Make runners executable and improve their shebangs	3 rokov pred
task.py	09240991cc Make model uploading use access token from authorizer (#7)	3 rokov pred
utils.py	f621362466 Make logging less verbose	3 rokov pred

Training DALL-E with volunteers from all over the Internet

This repository is a part of the NeurIPS 2021 demonstration "Training Transformers Together".

In this demo, we train a model similar to OpenAI DALL-E — a Transformer "language model" that generates images from text descriptions. Training happens collaboratively — volunteers from all over the Internet contribute to the training using hardware available to them. We use LAION-400M, the world's largest openly available image-text-pair dataset with 400 million samples. Our model is based on the dalle‑pytorch implementation by Phil Wang with a few tweaks to make it communication-efficient.

See details about how to join and how it works on our website.

README.md

Training DALL-E with volunteers from all over the Internet