Ran “sudo dpkg -i cuda-repo-ubuntu1504-7-5-local_7.5-18_amd64.deb” which works.
Ran “sudo apt-get update”
This fails with the following error:
E: Failed to fetch file:/var/cuda-repo-7-5-local/Release No Hash entry in Release file /var/lib/apt/lists/partial/_var_cuda-repo-7-5-local_Release which is considered strong enough for security purposes
E: Some index files failed to download. They have been ignored, or old ones used instead.
But installing the RC 8 version causes problems when later trying to use CUDA with packages installed with conda: when I try to run the example from the first “Accelerated Computing” notbook locally, the cuda toolkit is not found for the task 3 (numba) example. To solve this one has to install the cudatoolkit package into conda, but that package is only available in version 7.5
Also the installation for tensorflow seems to require version 7.5.
If you want to use CUDA 7.5, follow the instructions in the linux install guide. That includes using a supported linux distro for CUDA 7.5 (they are listed in that document). Ubuntu 16.04 is not supported.
So it isn’t possible to use this with a current version of Ubuntu.
Well that is a frustrating disappointment right before I actually started using it. Will wait until Nvidia gets a bit more usable then.
I encountered the same problem. I noticed that the most recent release of CUDA7.5 is for Ubuntu 15.04, fro 28. April 2016. So we will probably have to wait another year before we get it for Ubuntu 16.04???
Is there really no other way than using the .deb file for that specific Ubuntu version?
I don’t think you’ll ever see a version of CUDA 7.5 that “officially” supports Ubuntu 16.04.
However CUDA 8RC “officially” supports Ubuntu 16.04 today, assuming you are using gcc 5.3.1. Again, this info is covered in the installation guide for linux for CUDA 8RC.
Certainly there are people who have combined CUDA 7.5 and Ubuntu 16.04 in a non-officially-supported way. One such description is here: