Closed JVPC0D3R closed 5 months ago
Hi, thanks for the suggestions. The most recent versions of docker images are tagged with v3.0.0
, since this is the latest release. We briefly used nightly tags, but no longer use them. Therefore, nightly_290823
is older compared to v3.0.0
.
The actual issue here is that cuda_v3.0.0
is missing... I will fix this within the day probably.
The image was correctly built, but for some reason never pushed successfully to the online repo. The updated one should be available in a few minutes (it is uploading). Please revert this change in order to proceed with the torchvision
tag update. Please also target the develop
branch.
I'm excited to collaborate on this project! Thank you @passalis
Thanks! I am going to fix the error in another branch (it is related to outdated copyright strings) and then you can go ahead and merge this.
Update Torchvision Version and Docker CUDA Tag in Installation Docs
Changes
torchvision
installation command to includecu116
to ensure compatibility with CUDA 11.6.cuda_nightly_290823
, ensuring users have access to the latest features and fixes.Justification
torchvision
is critical as it provides the necessary binaries compatible with CUDA 11.6 setup.These updates ensure that the installation guide remains current.