Open calvintwr opened 1 week ago
When resuming finetuning, I see that the CycleIterator is forwarded to the dataset where the iteration is to continue from:
https://github.com/Lightning-AI/litgpt/blob/f3343784bbd192490e2a70aa5ef75c52608b1d35/litgpt/finetune/full.py#L208-L219
However, for pretrain, this does not exist and the training seems to resume from the begining:
https://github.com/Lightning-AI/litgpt/blob/f3343784bbd192490e2a70aa5ef75c52608b1d35/litgpt/pretrain.py#L217-L271
Can I check in this case, it looks like when resuming, the pretraining will start from the first dataset, and not forwarded?
The pretraining code uses a stateful dataloader from LitData: https://github.com/Lightning-AI/litgpt/blob/f3343784bbd192490e2a70aa5ef75c52608b1d35/litgpt/pretrain.py#L192-L198
When resuming finetuning, I see that the CycleIterator is forwarded to the dataset where the iteration is to continue from:
https://github.com/Lightning-AI/litgpt/blob/f3343784bbd192490e2a70aa5ef75c52608b1d35/litgpt/finetune/full.py#L208-L219
However, for pretrain, this does not exist and the training seems to resume from the begining:
https://github.com/Lightning-AI/litgpt/blob/f3343784bbd192490e2a70aa5ef75c52608b1d35/litgpt/pretrain.py#L217-L271
Can I check in this case, it looks like when resuming, the pretraining will start from the first dataset, and not forwarded?