Torch2coreml multiprocessing

[rbourgeat]

Hello ! First of all thank you for this great repository !

I just tried for the first time the torch2coreml module and i don't know why only one CPU and 0% GPU is used when i convert the sdxl-turbo model... If it's normal then why not do multiprocessing ?

Command used:

python -m python_coreml_stable_diffusion.torch2coreml --convert-unet --convert-text-encoder --convert-vae-decoder --convert-safety-checker --model-version stabilityai/sdxl-turbo -o sdxl-turbo

Macbook info:

Apple M1 Max
32Go RAM

Logs: https://gist.github.com/rbourgeat/9b7523d412a6eaf0cafbd91827a61454