Open yejr0229 opened 3 months ago
I encountered the same problem,How long did your code run after modifying the parameters?
Same problem, I guess it's a problem about the diffusers version. I'd like to konw which version of diffusers you are using? @Quantuman134
I encountered the same problem,How long did your code run after modifying the parameters?
Just delete the 'tau' parameter and it works fine.
Thanks! it worked.
I deleted the 'tau' parameter and it did worked. But the speed of generating textures by running code on A100 server is very slow, approximately 1 hour. I would like to know how long did your code run?
I run this command 'python main.py --mesh_dir 'Assets/Nascar.obj' --save_dir 'Results' --text_prompt 'The next gen of nascar'' and get the following error: Traceback (most recent call last): File "/home/yejr/AIGC/TexPainter-main/main.py", line 97, in
main(args)
File "/home/yejr/AIGC/TexPainter-main/main.py", line 58, in main
tex_paint(mesh_data, tex_latent, tex_rgb, DM_params, sd, camera_params, epochs, learning_rate)
File "/home/yejr/AIGC/TexPainter-main/function.py", line 76, in tex_paint
img_latent_model = latent_denoise(diffusion_model, img_latent, depth_map, text_embeddings, time_step, guidance_scale)
File "/home/yejr/AIGC/TexPainter-main/function.py", line 140, in latent_denoise
img_latent_model = diffusion_model.latents_denoise_step(latents=img_latent, depth_map=depth_map,
File "/home/yejr/AIGC/TexPainter-main/stable_diffusion_depth.py", line 87, in latents_denoise_step
latents_model = self.scheduler.step(noise_pred, t, latents, eta=eta, tau=tau)
TypeError: step() got an unexpected keyword argument 'tau'
Seems like self.scheduler.step() didn't have 'tau' parameter, and I try to delete this 'tau' and run latents_model = self.scheduler.step(noise_pred, t, latents, eta=eta), it works, so I'd like to know how to modify the code?