Please someone confirm that my below assumption is correct or not?

vision
1

Assumption:
If you set the dataloader class shuffle=True, you can get random samples from ENTIRE dataset into your batches. I mean the batches will be created with random samples of the ENTIRE dataset insead of sequential ones.

2

@ptrblck_de can you answer please…

3

Yes, using shuffle=True will use a RandomSampler internally as seen here and which will shuffle all indices in the dataset.

1 Like