Port `RASampler` functionality to iterable datasets

Our current classification references use the “Repeated Augment Sampler” (RASampler) from #5051:

https://github.com/pytorch/vision/blob/d585f86d94f07a3bc083e48c6534d93a409cbcb2/references/classification/sampler.py#L7

Since after the revamp we will have iterable- rather than map-style datasets, samplers are no longer supported.

Given that the RASampler increases accuracy, we need to support the same functionality going forward. It can probably be achieved by appending a custom RepeatedAugmentIterDataPipe to the dataset graph, but we need to make sure it works correctly with the shuffling and sharding.

cc @pmeier @YosuaMichael @datumbox @vfdev-5 @bjuncek

[pmeier]

One thing that came to mind thinking about this is that we will be wasting quite some I/O resources when using this functionality. The prototype datasets will read the raw bytes and store them in a tensor for later decoding. Since we will throw away the majority of the samples using this technique, we wasted all these reads. I don’t know if that is significant since we only read and not decode, but we should watch out for this.

[datumbox]

I see what you mean. You mean that we decode the same thing over and over. That’s expected and I agree it was happening on the previous implementation.

For a moment I thought you meant that the new API would require us decoding the part of the dataset that we throw-away, which would obviously be a problem. Sorry for the confusion.