MrForExample/ComfyUI-AnimateAnyone-Evolved

TypeError: can't convert cuda:0 device type tensor to numpy.

zymox opened this issue · 3 comments

zymox commented

It strangely ran for me like two times (?) but now I only get the following error any time the AnimateAnyone node is reached:

Error occurred when executing [ComfyUI-3D] Animate Anyone Sampler:

can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

File "E:\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "E:\ComfyUI\execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "E:\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "E:\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\nodes.py", line 152, in animate_anyone
samples = diffuser(
File "C:\Dev\Python3.10\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "E:\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\src\models\main_diffuser.py", line 440, in __call__
latents = self.denoise_loop(
File "E:\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\src\models\main_diffuser.py", line 324, in denoise_loop
self.scheduler.set_timesteps(loop_steps, device=device)
File "C:\Dev\Python3.10\lib\site-packages\diffusers\schedulers\scheduling_euler_discrete.py", line 318, in set_timesteps
sigmas = np.array(((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5)
File "C:\Dev\Python3.10\lib\site-packages\torch\_tensor.py", line 970, in __array__
return self.numpy()
zymox commented

This seems to happen when I select any other sampler than DDIM or PNDM in the "sampler_scheduler_pairs" parameter of the node.

TomTomGit86 commented

same here

i cant seem to use some of the diffrent sampler methods in the the animate anyone smapler, mainly the DPM++ 2M Karas will result in this error:

Error occurred when executing [ComfyUI-3D] Animate Anyone Sampler:

can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\nodes.py", line 152, in animate_anyone
samples = diffuser(
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\python_embeded\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\src\models\main_diffuser.py", line 440, in call
latents = self.denoise_loop(
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\src\models\main_diffuser.py", line 324, in denoise_loop
self.scheduler.set_timesteps(loop_steps, device=device)
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\python_embeded\lib\site-packages\diffusers\schedulers\scheduling_dpmsolver_multistep.py", line 263, in set_timesteps
sigmas = np.array(((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5)
File "C:\Users\thoma\Documents\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\python_embeded\lib\site-packages\torch_tensor.py", line 970, in array
return self.numpy()

Close
Queue size: 0⚙️
Queue Prompt
CPU
0%
RAM
30%
GPU
6%
VRAM
38%
HDD
63%
ERROR
Extra options
Queue FrontView QueueView History
Save
Load
Refresh
Clipspace
Clear
Load Default
Manager
Share
TypeError 31 [ComfyUI-3D] Animate Anyone Sampler

TomTomGit86 commented

maybe you should change the title of this issue to something more inviting like "unable to use other samplers like DPM++ 2M Karas"

  • 👍1