Data Loader Memory Consumption

[Happy_NewYear (Happy)]

I am using data loader for loading images but after running this loop, my GPU memory is getting full but it should not be happening because I am converting feat back to cpu and then moving over.

...
vgg16.cuda()
for i, (input,target,path) in enumerate(train_loader):
	feat = vgg16(Variable(input).cuda()).cpu()
	tr_data.append(feat)
...

What can be the cause?

[Happy_NewYear (Happy)]

Solved.

Data loaders, memory issues and circular references

Could you share your data loader code? Yes, it should not happen. in fact that is the whole point of having a data loader. One possible reason I can think of - if you’re not scaling them to a fixed size while reading them, then it’s possible that a very large image makes your batch too large for your GPU. It might be useful to see how your GPU usage grows over time across iterations. If it grows with every iteration, there’s a problem. If it suddenly grows out of bounds, there’s a problem in …