Open jy2247 opened 6 months ago
What version of numpy is installed in the container? You can check this by importing numpy in python, and running numpy.version.version
. Only other issue I could maybe see is the driver version - I am using 545.36, but unsure why that would be a problem...
This numpy version:
after I rebuild, it break at 12/18
@jy2247 can you please post what command you are running to build the chrono
container?
I think is this one: atk dev -dbua -s dev --optionals gpus vnc
Can you try pulling the containers rather than building: first, stop the running sim container with the atk -d
option. Then run docker system prune
to remove any old images, and try running atk dev -ua -s chrono --optionals gpus vnc
. If that doesn't work, the only issues I can think of is with the NVIDIA driver version, or that you are on Windows I believe (correct?), in which case I would have to do a bit more testing...
@StefanCaldararu @AaronYoung5
Not urgent but seems I could reproduce the error on my machine (windows 11):
Is this Windows 11 or within wsl?
And you've pulled the latest, deleted old chrono images?
Is this Windows 11 or within wsl?
Within WSL. Should I try on windows 11 directly?
And you've pulled the latest, deleted old chrono images?
Yea I did that.
Have you tried the existing chrono docker image, like the one that thomas made? Does that work?
And what about outside a container? Does it work on these computers without docker?
Has this been solved? @harryzhang1018
Has this been solved? @harryzhang1018
Sorry about the late reply. I tried Thomas' chrono ros docker image and it works well.
With the same computer as you're seeing this issue with?
@AaronYoung5 The link to the docker image is: https://hub.docker.com/r/uwsbel/ros/tags. You might want to ask Thomas to give you access on docker script to create this image.
With the same computer as you're seeing this issue with?
Yes, the same one, my laptop.
I followed setup steps and I could able to get into the chrono container without problem. Inside container, I can run "nvidia-smi" without problem. However, I get the following error when I try to run the python demo. I tried cpp demo as well and it seems to give the same error.