Offer an alternative to Iterable Dataset that allows lazy loading and processing while skipping batches efficiently

### Feature request

I would like a way to resume training from a checkpoint without waiting for a very long time when using an iterable dataset.

### Motivation

I am training models on the speech-recognition task. I have very large datasets that I can't comfortably store on a disk and also quite computationally intensive audio processing to do. As a result I want to load data from my remote when it is needed and perform all processing on the fly.

I am currently using the iterable dataset feature of _datasets_. It does everything I need with one exception. My issue is that when resuming training at a step n, we have to download all the data and perform the processing of steps < n, just to get the iterable at the right step. In my case it takes almost as long as training for the same steps, which make resuming training from a checkpoint useless in practice.

I understand that the nature of iterators make it probably nearly impossible to quickly resume training.

I thought about a possible solution nonetheless :

I could in fact index my large dataset and make it a mapped dataset. Then I could use set_transform to perform the processing on the fly. Finally, if I'm not mistaken, the _accelerate_ package allows to [skip steps efficiently](https://github.com/huggingface/accelerate/blob/a73898027a211c3f6dc4460351b0ec246aa824aa/src/accelerate/data_loader.py#L827) for a mapped dataset.

Is it possible to lazily load samples of a mapped dataset ? I'm used to [dataset scripts](https://huggingface.co/docs/datasets/dataset_script), maybe something can be done there.
If not, I could do it using a plain _Pytorch_ dataset. Then I would need to convert it to a _datasets_' dataset to get all the features of _datasets_. Is it something possible ?

### Your contribution

I could provide a PR to allow lazy loading of mapped dataset or the conversion of a mapped _Pytorch_ dataset into a _Datasets_ dataset if you think it is an useful new feature.