John Pellman

@guillaumeaubert - Ah, my apologies. I thought @gaubert might have gotten a new Github account. Honestly, I thought that your gmvault-docker project + your shared name was a clear indication...

Some sort of shared governance of the project would indeed be prudent (even if @gaubert still intends to work on this and is just temporarily side-tracked). Ideally a CentOS-like situation...

My summer's unfortunately a little busy so I don't think I'd be a very good mentor, but if I find anyone who would be willing to mentor a ``PythonTool`` /...

I think one point that's being lost in this conversation is that containerization is not necessarily equivalent to packaging and versioning software or documenting a computational environment. Containerization, as it...

To add to my earlier comment, it's worth taking a look at the article [Why Docker is not the answer to reproducible research, and why Nix may be](https://github.com/WeAreWizards/blog/blob/d5611b00e238115a6904f5f9e9d8772161e46c8e/content/articles/reproducible-research-or-why-docker-is-not-the-answer.md), which I...

Just as a caveat for those trying to use @fuog 's solution: While `install-redhat.yml` has a task to install Linux kernel headers (see [here](https://github.com/NVIDIA/ansible-role-nvidia-driver/blob/8946eefeb5442761edef6853665074e1306391a0/tasks/install-redhat.yml#L6-L15)), `install-ubuntu.yml` does not have an equivalent...

I'm encountering the same issue as well- not even SIGKILL will work to stop a script using PyTorch. Honestly, if SIGKILL isn't working, I doubt that this issue is specific...

For quick reference, here's a set of tallies for some of the relevant software/version number pairings: ## *NVIDIA Driver Summary Count:* |Version|Count| |-------|-------| |410.79|1| |418.67|1| |418.87|2| ## *CUDA Summary Count:*...

I was able to replicate this issue using an environment identical to my earlier one except for the CUDA version, which was 9.2.88.

As another data point, the version of fastai exhibiting a similar issue that I linked to in my earlier comment would have also used CUDA 9.2 (based off a contemporary...