Are there other ways to save the relationship between loss and learning rate

Part 1 (2019)
1

I use fastai’s learn class to train on a remote cloud server.
Generally, we will adjust the learning rate based on the relationship between the learning rate and the loss.

learn.lr_find()
learn.recorder.plot()

However, for some reason, the remote server cannot return this relationship graph.

Are there other ways we can Save the relationship between them and draw locally?

2

learn.recorder.plot(return_fig=True)

3

yes, use return_fig=True ,and savefig as a png image.

Thank you very much