Is it possible to generate using CPU ?

[heiheiheibj]

hello.

I had some trouble with my graphics card though. when i switch the device to CPU

HBox(children=(FloatProgress(value=0.0, description='Batches', max=50.0, style=ProgressStyle(description_width='initial')), HTML(value='')))

Output() 0%| | 0/240 [00:00<?, ?it/s] Seed used: 896041566 Traceback (most recent call last): File "disco.py", line 2533, in do_run() File "disco.py", line 1338, in do_run for j, sample in enumerate(samples): File "D:\disco-diffusion/guided-diffusion\guided_diffusion\gaussian_diffusion.py", line 900, in ddim_sample_loop_progressive eta=eta, File "D:\disco-diffusion/guided-diffusion\guided_diffusion\gaussian_diffusion.py", line 671, in ddim_sample model_kwargs=model_kwargs, File "D:\disco-diffusion/guided-diffusion\guided_diffusion\respace.py", line 91, in p_mean_variance return super().p_mean_variance(self._wrap_model(model), *args, kwargs) File "D:\disco-diffusion/guided-diffusion\guided_diffusion\gaussian_diffusion.py", line 260, in p_mean_variance model_output = model(x, self._scale_timesteps(t), model_kwargs) File "D:\disco-diffusion/guided-diffusion\guided_diffusion\respace.py", line 128, in call return self.model(x, new_ts, kwargs) File "D:\ProgramData\Anaconda3\envs\tf\lib\site-packages\torch\nn\modules\module.py", line 1102, in _call_impl return forward_call(*input, *kwargs) File "D:\disco-diffusion/guided-diffusion\guided_diffusion\unet.py", line 656, in forward h = module(h, emb) File "D:\ProgramData\Anaconda3\envs\tf\lib\site-packages\torch\nn\modules\module.py", line 1102, in _call_impl return forward_call(input, kwargs) File "D:\disco-diffusion/guided-diffusion\guided_diffusion\unet.py", line 77, in forward x = layer(x) File "D:\ProgramData\Anaconda3\envs\tf\lib\site-packages\torch\nn\modules\module.py", line 1102, in _call_impl return forward_call(*input, **kwargs) File "D:\ProgramData\Anaconda3\envs\tf\lib\site-packages\torch\nn\modules\conv.py", line 446, in forward return self._conv_forward(input, self.weight, self.bias) File "D:\ProgramData\Anaconda3\envs\tf\lib\site-packages\torch\nn\modules\conv.py", line 443, in _conv_forward self.padding, self.dilation, self.groups) RuntimeError: "unfolded2d_copy" not implemented for 'Half'

What should I do ? thx

[entmike]

Known issue apparently. Reported by others on Discord. Have not seen anyone able to fix yet.

[OodavidsinoO]

I made it worked on the CPU. Here's my repo. https://github.com/OodavidsinoO/disco-diffusion

[heiheiheibj]

OK.thx

[aletts]

The option to use the CPU is now in official Disco Diffusion 5.2 too.

On Sun., May 15, 2022, 11:43 a.m. heiheiheibj @.***> wrote:

OK.thx

— Reply to this email directly, view it on GitHub https://github.com/alembics/disco-diffusion/issues/50#issuecomment-1126966875, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABCZGPTKJKR2CXDRJ6UI7LVKELTFANCNFSM5TM4H6HA . You are receiving this because you are subscribed to this thread.Message ID: @.***>