Spaces:

vinthony

/

SadTalker

like 1.19k

Running on A10G

My configuration:

• Tencent Cloud 腾讯云
• NVIDIA GV100GL [Tesla V100 SXM2 32GB]
• Ubuntu 22.04.3 LTS
• Python 3.8.17

I have followed https://github.com/OpenTalker/SadTalker#linux

After running pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 torchaudio==0.12.1 --extra-index-url https://download.pytorch.org/whl/cu113 I got:

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
accelerate 0.21.0 requires pyyaml, which is not installed.
aiohttp 3.8.4 requires attrs>=17.3.0, which is not installed.
altair 5.0.1 requires jinja2, which is not installed.
altair 5.0.1 requires jsonschema>=3.0, which is not installed.
datasets 2.13.0 requires pyyaml>=5.1, which is not installed.
gradio 3.40.1 requires jinja2<4.0, which is not installed.
gradio 3.40.1 requires pyyaml<7.0,>=5.0, which is not installed.
huggingface-hub 0.16.4 requires pyyaml>=5.1, which is not installed.
modelscope 1.8.1 requires attrs, which is not installed.
modelscope 1.8.1 requires pyyaml, which is not installed.
modelscope 1.8.1 requires simplejson>=3.3.0, which is not installed.
oss2 2.18.1 requires six, which is not installed.
peft 0.4.0 requires pyyaml, which is not installed.
pydeck 0.8.1b0 requires jinja2>=2.10.1, which is not installed.
streamlit 1.24.1 requires blinker<2,>=1.0.0, which is not installed.
streamlit 1.24.1 requires click<9,>=7.0, which is not installed.
transformers 4.31.0 requires pyyaml>=5.1, which is not installed.
uvicorn 0.23.2 requires click>=7.0, which is not installed.
modelscope 1.8.1 requires numpy<=1.22.0, but you have numpy 1.24.4 which is incompatible.
peft 0.4.0 requires torch>=1.13.0, but you have torch 1.12.1+cu113 which is incompatible.
streamlit 1.24.1 requires pillow<10,>=6.2.0, but you have pillow 10.0.0 which is incompatible.
xformers 0.0.20 requires torch==2.0.1, but you have torch 1.12.1+cu113 which is incompatible.

Any ideas? Thank you!