Clarification on Default Training Parameters in train_ood_det_in100.py

[chalengr]

Hello,

I am currently working with the train_ood_det_in100.py script and noticed that the default number of training epochs is set to 20:

parser.add_argument('--epochs', '-e', type=int, default=20, help='Number of epochs to train.')

I would like to confirm whether this default value aligns with the parameters used in your paper. In particular, I am interested in understanding if the 20 epochs are sufficient to replicate the results reported in your publication.

Thank you for your time.