Dataset weightage bug

[yhl48]

It doesn’t look like this line is doing what’s intended based on the comment, all datasets are given equal weightage here

litdata/src/litdata/streaming/combined.py

Line 74 in d5eff39

self._weights = [1 / float(num_datasets)] * num_datasets

cc: @tchaton

[yhl48]

For context, I started looking into this code because I consistently got periodic sine-like validation loss when combining two datasets, but the issue disappeared when I manually shuffled and combined them into a single dataset. The two datasets have a 1:3 size ratio.

I am not sure if L74 is a bug, and if it is indeed a bug, I am not sure if correcting that would solve the validation loss issue. This could be due to the two datasets that I have being very different, prompting me to think about if there is a good sampling strategy that is universally acceptable. Anyways, I thought it would be good to have a discussion here 🙂.

[tchaton]

Oh yes, good catch, we should take the length of each dataset normalized by the total.

dataset_lens = [len(d) for d in datasets]
total = sum(dataset_lens)
self._weights = [l / total for l in dataset_lens]