Learning rate finder shows #na# under val_loss

It doesn’t mean anything. For the LR finder, the validation loss is not calculated, hence #na#.

4 Likes