Log current lr in `StandardModel`

graphnet-team / graphnet

A Deep learning library for neutrino telescopes

https://graphnet-team.github.io/graphnet/

Apache License 2.0

85 stars 86 forks source link

Closed RasmusOrsoe closed 5 months ago

RasmusOrsoe commented 5 months ago

This PR changes the following in graphnet.models.StandardModel:

Adds the global optimizer learning rate as a logged quantity, which makes it appear both in wandb and in the ProgressBar.
Sets on_step=True for the logging of training loss.

This change makes it easier to monitor the progress of the training from the progressbar information alone.