Datasets:

facebook

/

wiki_dpr

like 33

Follow

AI at Meta 3,337

Reported by @kiehls90: https://github.com/huggingface/datasets/issues/6620

I'm trying to run a rag example, and the dataset is wiki_dpr.
wiki_dpr download and extracting have been completed successfully.
However, at the generating train split stage, an error from wiki_dpr.py keeps popping up.

Especially in "_generate_examples" :

1. The following error occurs in the line id, text, title = line.strip().split("\t")
   ValueError: not enough values ​​to unpack (expected 3, got 2)
   -> This part handles exceptions so that even if an error occurs, it passes.

2. ID mismatch between lines {id} and vector {vec_id}
   This error seems to occur at the line " assert int(id) == int(vec_id),".
   After I handled the exception in the split error, generating train split progressed to 80%, but an id mismatch error occurred at about the 16200000th vector id.

Debugging is even more difficult because it takes a long time to download and split wiki_dpr. I need help. thank you in advance!!

Steps to reproduce the bug

Occurs in the generating train split step when running the rag example in the transformers repository.
Specifically, it is an error in wiki_dpr.py.