Closed Z0rx closed 1 year ago
I couldn't replicate that. Could you provide more context? My prompt and negative prompt are ~150 tokens long, and I'm using: python: 3.10.6 • torch: 1.13.1+cu117 • xformers: 0.0.16rc425 • gradio: 3.23.0 • A1111 commit: 22bcc7be • controlnet: v1.1.173 • NPW commit: 72a9348
I think this is caused by a fresh version of controlnet (the error appeared near the version ControlNet v1.1.180+). The problems started after the controlnet update, as it seemed to me.
I couldn't replicate that. Could you provide more context? My prompt and negative prompt are ~150 tokens long, and I'm using: python: 3.10.6 • torch: 1.13.1+cu117 • xformers: 0.0.16rc425 • gradio: 3.23.0 • A1111 commit: 22bcc7be • controlnet: v1.1.173 • NPW commit: 72a9348
Update Controlnet :)
Should be fixed. At least it works! Please update and test to see if it works as expected, and let me know if I should close the issue.
Seems to be no problem now. Thank you.
Fixed thanks!
Hey,
Today i updated controlnet and NPM , so ill started to use controlnet. If you set NPM above 1.00 wight the script drops an error.
" File "C:\Users\datenschutzduhrnsn\Desktop\ai\stable-diffusion-webui\extensions\stable-diffusion-NPW\scripts\npw.py", line 90, in denoiser_callback new_uncond = torch.lerp(empty_uncond_concat, uncond, self.weight) RuntimeError: The size of tensor a (1463) must match the size of tensor b (1464) at non-singleton dimension 1 "