I am trying to use this imagenet script: https://github.com/pytorch/examples/blob/master/imagenet/main.py on a pretrained mobilenetV2 model using my own custom dataset. Unfortunately, I am not having any luck. I am still new to Pytorch, having just learned about it last month (June 2020). Is there a link with a tutorial or example to help better understand the code and how to modify it to fit my needs?
I would recommend to check out the tutorials first and in particular this transfer learning tutorial.
This should explain a lot about the underlying Dataset and DataLoader usage as well as about the general training structure. 