Philippe Manzano

I'm trying to understand what I did wrong. Just to be clear, after downloading mnet4, this one is using mnet4, not using Docker from master branch (my other ticket just...

`pip install --upgrade nvidia-cudnn-cu12` doesn't work because of torch 2.5.1 I wonder if the mismatch is: Cuda 12.6 requires latest packages, but latest Torch is still on 'older' package versions.

Thank you very much @AmmarkoV I undestand about OpenGL, you're right. > As you see from your output tensorflow does not use GPU: Actually Tesnorflow sees the GPU: `>>> import...

> If the issue is processing many files rather than a single one, you'll probably get faster overall performance if you disable Rhubarb's multithreading altogether and run several instances in...

Also getting the same issue, any love from devs?

Same error here using CUDA 12.6, can't downgrade. Would it be possible to make a docker container that use CUDA 11 while accessing the webcam on host? Since it's a...

Maybe I need to install a brand new python=3.10 environment, then pip install -e . :smile:

Sorry, I thought you knew about the snap package. BTW, his 5.3 is broken on my Ubuntu 22.04, menu for `instances` and `models` are greyed, so it's a no go...

I did, edge is 5.3 

Nope, as mentioned above `his 5.3 is broken on my Ubuntu 22.04, menu for instances and models are greyed`