Missing labels ValueError exception
#10
by akoumpa - opened
Hi folks,
Thank you for sharing your model with the open-source community.
While trying this model I found that if I attempt to train the model without passing the labels argument to the forward function I'm hitting a ValueError exception here.
I wonder if it would make sense to remove it and whether that would break anything in the model.
For what it's worth, my recommendation would be to opt for a one-time warning message instead of the ValueError, but I'm also curious to learn your thoughts.
Looking forward to your response, best.