Open yhyu13 opened 6 years ago
@yhyu13
I believe it is because the saved model or memories were done using CPU or GPU and you tried to load them with the other type. Deleting them should fix it like you said. Probably that would be a good reason to .gitignore the checkpoints folder, I'll try to add that at some point.
Hi, @jprothero
I was testing your newest tweaking but I found an error about cpu vs gpu:
The main error is
I guess the model you committed was trained on cpu alone, right? And when I load the model onto gpu, it expect different type of variable? Do you know any solution to fix this kind of issue in pytorch? (Delete your model and memory.* solve this problem, but I am still confused about cpu vs gpu in pytorch a bit)