moyapchen commented 3 years ago

Environment info

transformers version: 4.3.2
Platform: Linux-5.4.0-52-generic-x86_64-with-debian-bullseye-sid
Python version: 3.6.12
PyTorch version (GPU?): 1.7.1+cu110 (True)
Tensorflow version (GPU?): not installed (NA)
Using GPU in script?: N/A
Using distributed or parallel set-up in script?: nope

Who can help

Models:

rag: @patrickvonplaten, @lhoestq

Information

Model I am using (Bert, XLNet ...): RAG

The problem arises when using:

[ x ] the official example scripts: (give details below)
[ ] my own modified scripts: (give details below)

The tasks I am working on is:

[ ] an official GLUE/SQUaD task: (give the name)
[x ] my own task or dataset: (give details below)

To reproduce

Steps to reproduce the behavior:

Run any of the scripts in the examples on https://huggingface.co/transformers/model_doc/rag.html#overview , ex.

from transformers import RagTokenizer, RagRetriever, RagModel
import torch
tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-base")
retriever = RagRetriever.from_pretrained("facebook/rag-token-base", index_name="exact", use_dummy_dataset=True)
# initialize with RagRetriever to do everything in one forward call
model = RagModel.from_pretrained("facebook/rag-token-base", retriever=retriever)
input_dict = tokenizer.prepare_seq2seq_batch("How many people live in Paris?", "In Paris, there are 10 million people.", return_tensors="pt")
input_ids = input_dict["input_ids"]
outputs = model(input_ids=input_ids)

Get an error on https://github.com/huggingface/transformers/blob/master/src/transformers/models/rag/tokenization_rag.py#L77 about how super() does not have prepare_seq2seq_batch()
- Indeed, looking at the relevant file, RagTokenizer does not inherit from any other class.

Expected behavior

RAG works properly.

Note that if I copy/paste the code in the file prior to https://github.com/huggingface/transformers/pull/9524 , it works fine. CC: @sgugger of that change.

lhoestq commented 3 years ago

Hi ! Thanks for reporting

10167 should fix this issue