Retrieve LR

[bhack]

What is the best way to retrieve the LR for this optimizer at each train step to plot an LR curve without the LR scheduler?

Is it similar to: https://discuss.pytorch.org/t/get-current-lr-of-optimizer-with-adaptive-lr/24851/2

[adefazio]

The LR doesn't change during optimization. It achieves convergence by averaging rather than decreasing the learning rate.