lehduong/torch-warmup-lr

update _last_lr in step()

Opened this issue · 0 comments

First of all, thanks for your nice implementation of Warmup wrapper.

I found one tiny thing to improve.

By adding self._last_lr = [group["lr"] for group in self.optimizer.param_groups] at the end of step() function, we could get right value of lr when we call scheduler.get_last_lr()[0] to check lr value.