bug fix (evaluation.mdx): language specification required to run the prediction code.
bug fix (fine-tuning.mdx):
Seq2SeqTrainingArguments requires dataloader_num_workers and dataloader_prefetch_factor parameters; otherwise, it throws errors due to a bug in some versions of the PyTorch data loader. It's safe to include these two parameters.
bug fix (evaluation.mdx): language specification required to run the prediction code.
bug fix (fine-tuning.mdx):
Seq2SeqTrainingArguments requires dataloader_num_workers and dataloader_prefetch_factor parameters; otherwise, it throws errors due to a bug in some versions of the PyTorch data loader. It's safe to include these two parameters.