Karel

Hi, sorry for late reply. We had used the DUT-TR dataset for training.

Hi @arjunsankar, Did you try `--num_workers=0` ? I recommend coordinating both num_workers and batch_size

Hi, did you change the num of batch size?

To compute the loss, you should set the values of label to 0, not nan. Since you set the values to nan, the loss.backward error occurred.