Nathaniel Simard

I resolved the issue by adding the lib directory of the virtualenv: ```bash export LD_LIBRARY_PATH=/home/nathaniel/.pyenv/versions/3.9.16/lib:$LD_LIBRARY_PATH ``` It took me too much time to figure this out, this could be better...

It might be an issue with pyenv, I expected it to handle such things. I'm not sure how extensions in C/C++ behave in that situation, but if they require python-dev,...

Just the RAM, nothing fancy :)

Yes no problem, you can submit a PR with only one metric if you want!

You can add metrics with any machine. To understand them, you can probably read some articles first like https://neptune.ai/blog/performance-metrics-in-machine-learning-complete-guide and then maybe use wikipedia for each individual metric you want...

Hi, I cannot reproduce the bug on Gnome-3.38 with the new fixes on #31. What is your setup exactly?

I'm not yet on Gnome 44.2, so there might be problems, you have more chance to make it work by using the git version. There are issue when deploying to...

It would be possible to have selectors for workspaces, but I'm not sure to understand the benefits of that. Because each window of every workspaces of every monitors can be...

@cwfitzgerald, we are indeed calling `get_bind_group_layout` not just at every frame, but for every compute kernel that we execute (can be thousands or more each second). See here: https://github.com/tracel-ai/burn/blob/3eab14160875ddaa1d0527247c09d6f37f8c75c7/burn-wgpu/src/compute/server.rs#L333 We...

Just to clarify, instead of calling `get_bind_group_layout` on the cached `ComputePipeline` when we want to execute it with different buffers, we should cache the `BindGroupLayout` along with the `ComputePipeline` when...