You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is docker image nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04 more suitable than nvidia/cuda:12.0.0-runtime-ubuntu22.04 for faster-whisper's cudnn requirements?
#998
Open
yasu-kondo opened this issue
Sep 10, 2024
· 0 comments
I'm playing with faster-whisper on Docker environment.
Let me suggest tiny modification on README.md.
I've struggled with following error message while I played with wrong environment, like cudnn9 instead of cudnn8.
Could not load library libcudnn_ops_infer.so.8. Error: libcudnn_ops_infer.so.8: cannot open shared object file: No such file or directory
Suggestion
modify README.md especially Use Docker section.
Suggestion
### Use Docker
The libraries (cuBLAS, cuDNN) are installed in these official NVIDIA CUDA Docker images: nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu20.04 or nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04.
Current
### Use Docker
The libraries (cuBLAS, cuDNN) are installed in these official NVIDIA CUDA Docker images: nvidia/cuda:12.0.0-runtime-ubuntu20.04 or nvidia/cuda:12.0.0-runtime-ubuntu22.04.
Some nvidia/cuda images don't have cudnn
faster-whisper needs cudnn package, but some docker image suggested on README.md doesn't have it.
$ docker run nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04 dpkg -l | grep cudnn
hi libcudnn8 8.9.6.50-1+cuda12.2 amd64 cuDNN runtime libraries
$ docker run nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04 dpkg -l | grep cublas
hi libcublas-12-2 12.2.5.6-1 amd64 CUBLAS native runtime libraries
$ docker run nvidia/cuda:12.0.0-runtime-ubuntu22.04 dpkg -l | grep cudnn # NO OUTPUT means that this image doesn't have `cudnn` package.
$ docker run nvidia/cuda:12.0.0-runtime-ubuntu22.04 dpkg -l | grep cublas
hi libcublas-12-0 12.0.1.189-1 amd64 CUBLAS native runtime libraries
I'm happy to make PR if this suggesion looks good ;)
The text was updated successfully, but these errors were encountered:
I'm playing with faster-whisper on Docker environment.
Let me suggest tiny modification on README.md.
I've struggled with following error message while I played with wrong environment, like
cudnn9
instead ofcudnn8
.Suggestion
modify README.md especially
Use Docker
section.Some nvidia/cuda images don't have
cudnn
faster-whisper
needscudnn
package, but some docker image suggested on README.md doesn't have it.I'm happy to make PR if this suggesion looks good ;)
The text was updated successfully, but these errors were encountered: