Missing config.json file

[DiogoBerna]

I am trying to run the following code:

import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
from datasets import load_dataset

device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

model_id = "distil-whisper/distil-large-v2"

model = AutoModelForSpeechSeq2Seq.from_pretrained(
    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
)

But I am getting the following error:

OSError: distil-whisper/distil-large-v2 does not appear to have a file named config.json. Checkout 'https://huggingface.co/distil-whisper/distil-large-v2/main' for available files.

Am I doing something wrong?

Btw, the distil-whisper instructions says Distil-Whisper is supported in Hugging Face Transformers from version 4.35 onwards, but I wasn't able to find that version: https://pypi.org/project/transformers/#history

[AnkushMalaker]

Model not uploaded yet I believe. The repo that the model links to, at this moment infact doesn't have config.json as the error suggests.

[kristiankielhofner]

It's fairly common for "bleeding edge" stuff to need to install transformers from git main:

pip install git+https://github.com/huggingface/transformers

This shows version transformers-4.35.0.dev0 currently.

[DiogoBerna]

Model not uploaded yet I believe. The repo that the model links to, at this moment infact doesn't have config.json as the error suggests.

That makes sense. It seems like they have updated this repo here before updating the model

[sanchit-gandhi]

Hey @DiogoVidevince - the models will be available in 2-3 hours time! Thanks for your patience!

[sanchit-gandhi]

Models are up! Please see the usage section of the README for detailed code examples. Feel free to open a new issue if you experience any problems!