yuval-alaluf/stylegan3-editing

Can I run the repo without GPU?

zydjohnHotmail opened this issue · 3 comments

Hello:
I want to use the repo to generate some images. However, my PC has NVIDIA Geforce gt 1030 graphic card. This graphic card has no ability to run any CUDA operation, but installing CUDA package is OK. But my PC has 32-core CPU with 256GB RAM, so I think CPU is powerful enough to do almost all the jobs.
I don't want to train any model, I just want to use pre-trained (.PT) models to generate some human faces or animal images.
Let me know in which python file, I can change so that the image predition use only CPU not GPU.
Thanks,

you can start with a 3060 12gb
https://github.com/NVlabs/stylegan3#requirements
#12
(CPU may be possible with some modifications, but is not inherently supported)
https://stackoverflow.com/questions/51730880/where-do-i-get-a-cpu-only-version-of-pytorch
You will need to create a fork from this repo

CPU may be possible with some modifications, but is not inherently supported.

Reference: https://github.com/yuval-alaluf/stylegan3-editing#prerequisites

Hello: I want to use the repo to generate some images. However, my PC has NVIDIA Geforce gt 1030 graphic card. This graphic card has no ability to run any CUDA operation, but installing CUDA package is OK. But my PC has 32-core CPU with 256GB RAM, so I think CPU is powerful enough to do almost all the jobs. I don't want to train any model, I just want to use pre-trained (.PT) models to generate some human faces or animal images. Let me know in which python file, I can change so that the image predition use only CPU not GPU. Thanks,

how did you solve this issue?? I have same problem and I want to run only on CPU, not GPU