huggingface / transformers

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
https://huggingface.co/transformers
Apache License 2.0
133.51k stars 26.67k forks source link

[Multiple errors] Maintainablity video-text-to-text in transformers? #34060

Open SoyGema opened 1 week ago

SoyGema commented 1 week ago

System Info

transformers==4.45.2

Who can help?

No response

Information

Tasks

Reproduction

Context

Hello there! Im going through the video-text-to-text content and I am experiencing a series of issues with broken links, unexistent content and errors. As they all come from the task page initially, decided to list them here as a hole possible maintenance issue, but please feel free to decompose if you find that this works best for you

Listed issues in Video-text-to-text page

  1. [Link broken] Unaccesible Codelab . When trying to access to both the colab (both on pytorch and tensorflow ) and amazon SageMaker Studio Lab, seems to point to an existing path , probably you changed it , and currently the notebook is not under your examples (Images above) Captura de pantalla 2024-10-10 a las 10 31 57 1

![Uploading Captura de pantalla 2024-10-10 a las 11.02.17.png…]()

  1. [Error in code] Sent a quick PR for this, missing an import #34058

  2. [Link broken] link at the end of the page , in the text To learn more about chat templates and token streaming for video-text-to-text models, refer to the [image-text-to-text](https://huggingface.co/docs/transformers/image_text_to_text) task guide because these models work similarly. , when clicking , it goes to 2 links resulting in a 404 error.

    Captura de pantalla 2024-10-10 a las 10 32 11
Captura de pantalla 2024-10-10 a las 10 32 22

Is this somehow similar to what happened in #24497 ??

Thanks for having a look to this ! Keep up the good work and have a nice day!

Expected behavior

Links working Code working

LysandreJik commented 6 days ago

cc @merveenoyan maybe? :)