Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Tests/docker (#1573)
* devel image * try parallel * new image
J
Jirka Borovec committed
0b22b64a1097e2a4b2027a4d90ff8b2341f82237
Parent: e977d1c
Committed by GitHub <noreply@github.com>
on 4/23/2020, 4:52:59 PM