search

X-PLUG / mPLUG-Owl

mPLUG-Owl: The Powerful Multi-modal Large Language Model Family
https://www.modelscope.cn/studios/damo/mPLUG-Owl
MIT License
2.25k stars 171 forks source link

Conflicting requirements #18

Closed joehoover closed 1 year ago

joehoover commented 1 year ago

There are conflicts between the dependencies specified in the README and env.yaml.

In the README, torchvision is not specified and torch is pinned at 1.13.1. Further, in env.yaml peft is not listed as a dependency.

PyTorch=1.13.1 (1.13.1 is required by the peft)

However, when I tried to run the demo like:

python -m server_mplug.owl_demo --debug --port 6363 --checkpoint_path 'your checkpoint path' --tokenizer_path 'your tokenizer path'

I got an import error for torchvision.

I then referred to env.yaml for dependencies, but there torch is pinned at 1.12.1

  - pytorch=1.12.1=py3.10_cuda11.3_cudnn8.3.2_0

This conflicts with the version specified in the README.

MAGAer13 commented 1 year ago

Thanks for your notice. We have updated now. By the way, we would clean up the code in the next version which no longer rely on apex dependency. Stay Tuned.

joehoover commented 1 year ago

Wow, thanks for the fast response!

I might just wait for the next version. Do you have an ETA for that?

Unfortunately, I ran into another issue with dependency conflicts.

torchvision==0.13.1 is not compatible with torch==1.13.1.

What version of torchvision would you recommend?

MAGAer13 commented 1 year ago

You might use the following script here to install the torch and torchvision for your reference. conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia

For the next version, we aim to natively support the huggingface library! Besides, we are working on multi-lingual support of that. We would release it as soon as the training is done.

MAGAer13 commented 1 year ago

If you have any further question or need help, we can set up a slack / discord channel to help you. 😊

MAGAer13 commented 1 year ago

Wow, thanks for the fast response!

I might just wait for the next version. Do you have an ETA for that?

Unfortunately, I ran into another issue with dependency conflicts.

torchvision==0.13.1 is not compatible with torch==1.13.1.

What version of torchvision would you recommend?

Hi, just quick notice for the next version have been released with much easier usage!