Michael New
Michael New
@karabaralex no, that warning almost certainly means you aren't running on the GPU. For me it's about a 50x speedup on the GPU vs CPU, so it should be pretty...
It is rocm. PyTorch just refers to it as cuda.
I was getting this error when trying to run with the `OPENAI_API_KEY` environment var. Switching to using `api_key_cmd` made it go away. If you just want to test you can...
FWIW, this is my working requirements.txt file: ```txt scikit-image numpy matplotlib argparse Pillow IPython torch ``` And to run on AMD, add the line: `--extra-index-url https://download.pytorch.org/whl/rocm5.7` above `torch`
Should be fixed with #90
Was anyone able to sort this out? Currently stuck on the same issue on Arch Linux. EDIT: Scratch that, I just had to change the imports to `import x11/[x, xlib]`...
You should be able to change the number of prints. Either just type a number into the box or use the little up/down arrows that appear when you mouse over....
I think the best way to do that would be to use something like [this](https://plugins.octoprint.org/plugins/gcodesystemcommands/). Just put a gcode hook in the bed clearing script and you should be able...
Try putting `M400` at the start of the script. The script is just prepended to the "after print job completes" script, so I'm (mostly) positive that the gcode isn't actually...
I'm still pretty sure that the gcode isn't actually arriving out of order, but if you want to check the terminal output in OctoPrint you should be able to tell...