GitLab

save best val loss in checkpoint and also print best so far

this way when training continues from an existing checkpoint, we dont immediately override checkpoint_best with a worse loss