Fix LR schedule handling for low-bit optimizers (#736)
* update benchmark script * try CPU Tensor lr * update * consolidate adam * remove unnecessary requires_grad in subclass * update README
T
Thien Tran committed
aacaf9b3e13461d1b2ff39ca2bf23aedfda259f7
Parent: d26bcca
Committed by GitHub <noreply@github.com>
on 8/23/2024, 3:43:00 PM