AIForce/train-dalle-together @ c61c61b20d956bfb3d43f87ebf796cbbb6b826b9

Нема описа

Aleksandr Borzunov c61c61b20d Use t5-small tokenizer		пре 3 година
lib	20e2a3aab2 Use dalle-pytorch instead of LeanAlbert	пре 3 година
.gitignore	72fc0bcdb7 Initial commit (ru-max branch without private code)	пре 3 година
README.md	72fc0bcdb7 Initial commit (ru-max branch without private code)	пре 3 година
arguments.py	c61c61b20d Use t5-small tokenizer	пре 3 година
callback.py	72fc0bcdb7 Initial commit (ru-max branch without private code)	пре 3 година
data.py	c61c61b20d Use t5-small tokenizer	пре 3 година
huggingface_auth.py	72fc0bcdb7 Initial commit (ru-max branch without private code)	пре 3 година
requirements.txt	72fc0bcdb7 Initial commit (ru-max branch without private code)	пре 3 година
run_aux_peer.py	f621362466 Make logging less verbose	пре 3 година
run_trainer.py	380a5f3231 Set num threads to <= 4	пре 3 година
run_trainer_tpu.py	380a5f3231 Set num threads to <= 4	пре 3 година
task.py	c61c61b20d Use t5-small tokenizer	пре 3 година
utils.py	f621362466 Make logging less verbose	пре 3 година

TODO write a setup instruction

Note: You might want to have several initial peers so that if one dies,

new workers still can join the collaboration via alive initial peers' addresses.
Specify initial_peers argument for that purpose