Reduce LR on plateau but not increase about nequip HOT 8 CLOSED

I think there is an option for this in PyTorch, though I ahven't looked super closely. See is_better in https://pytorch.org/docs/1.11/_modules/torch/optim/lr_scheduler.html#ReduceLROnPlateau.

Something to do with setting threshold?

You can specify these options in the YAML with lr_scheduler_* keys.

from nequip.

That's exactly what I need. These options aren't listed in https://github.com/mir-group/nequip/blob/main/configs/full.yaml. Do we need code changes to support them? Or will it automatically translate any lr_scheduler_X option into the X argument of the scheduler?

from nequip.

It should do it automatically; the configuration system is built to automatically propagate options from their hierarchical prefixes into the right objects. You can confirm this by running briefly in verbose: debug mode which explicitly logs the mapping from input keys to various objects being built to see if the right values are set when ReduceLROnPlateau gets instantiated (just grep ReduceLROnPlateau in the log).

from nequip.

This should already be implemented, you can set a delta on the improvement, see here and here. Let us know if you have questions re this.

from nequip.

This should already be implemented, you can set a delta on the improvement, see here and here.

That refers to early stopping, not to reducing the learning rate.

from nequip.

Oh my bad, of course, misread.

from nequip.

Setting lr_scheduler_threshold works perfectly. Thanks! Can we document in configs/full.yaml that you can specify any argument to the scheduler?

from nequip.

Done 👍

from nequip.