How to load complete UCF101 dataset?

vision
#1

Hi!
I am using the UCF101 dataset ( torchvision.datasets. UCF101). I want to load the complete dataset to make custom splits. However, I only see the boolean ‘train’ as an argument? Is there a way to create a DataLoader for the entire dataset?

Thanks!

#2

perhaps, you can load both train/test and merge (like concat) into one? since we can do split, I guess we should be able to merge too.

#3

I guess I’ll try that. I was hoping there would be a better way to load the entire dataset instead of combining the train/test splits and the folds.

Thanks! I’ll get back after I try it