Can't get it to run on Apple silicon.

[elchupacabrinski]

I always get the following error:
RuntimeError: slow_conv2d_forward_mps: input(device='cpu') and weight(device=mps:0') must be on the same device

How can stuff even be on different devices without a GPU? :D

[baicuya]

I got same issue:RuntimeError: slow_conv2d_forward_mps: input(device='cpu') and weight(device=mps:0') must be on the same device

[baicuya]

I have solved it ,you can add code /stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/openpose/body.py line48:
if torch.backends.mps.is_available():
mps_device = torch.device("mps:0")
data = data.to(mps_device)

[sonluong1987]

Thank you!
If you use openpose full then add the above code to these lines as well:

hand.py line 54

    with torch.no_grad():  
        if torch.backends.mps.is_available():  
            mps_device = torch.device("mps:0")  
            data = data.to(mps_device)  
        output = self.model(data).cpu().numpy()  

face.py line 341

  with torch.no_grad():
      if torch.backends.mps.is_available():
          mps_device = torch.device("mps:0")
          x_data = x_data.to(mps_device)
      hs = self.model(x_data[None, ...])
      heatmaps = F.interpolate(
          hs[-1],
          (H, W),
          mode='bilinear', align_corners=True).cpu().numpy()[0]

[pandora523]

mine can't update media-pipe
I did install open-cv but it was uptodate

Python 3.10.9 (main, Jan 11 2023, 09:18:18) [Clang 14.0.6 ]
Commit hash: 22bcc7be428c94e9408f589966c2040187245d81
Installing requirements for Web UI
Installing sd-webui-controlnet requirement: mediapipe==0.9.1.0
Couldn't install sd-webui-controlnet requirement: mediapipe==0.9.1.0.
Command: "/Users/rei/stable-diffusion-webui/venv-torch-nightly/bin/python3.10" -m pip install mediapipe==0.9.1.0 --prefer-binary
Error code: 1
stdout:
stderr: ERROR: Could not find a version that satisfies the requirement mediapipe==0.9.1.0 (from versions: 0.9.3.0)
ERROR: No matching distribution found for mediapipe==0.9.1.0

[notice] A new release of pip available: 22.3.1 -> 23.1
[notice] To update, run: pip install --upgrade pip

Warning: Failed to install mediapipe==0.9.1.0, some preprocessors may not work.
Installing sd-webui-controlnet requirement: opencv-python

[pswai]

im not sure if I did it right

im now getting this on Terminal. ModuleNotFoundError: No module named 'dumpy'

Your body.py is correct, but not the other two files.

For hand.py, the problem is the indentation (number of whitespaces). You need to keep the if 4 spaces away from the with, telling Python that that part of the code is "inside" the with. The same thing with line 55 and 56. Here's a screenshot of mine.

Similar for face.py:

Note that the code being pasted to all 3 files is the same:

if torch.backends.mps.is_available():
    mps_device = torch.device("mps:0")
    x_data = x_data.to(mps_device)

Just need to get the spacing right!

[pandora523]

I think it works!

About this one idid rename the the line 2 from dumpy to numpy. and have no idea what im doing just saw the code saying numpy so I did renamed it and it works!? lol. im not sure if this solves all my problems but that was really helpful !😆
God Blessed You By The Computer Gods!!!

File "/Users/lei/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/openpose/body.py", line 2, in
import dumpy as np
ModuleNotFoundError: No module named 'dumpy'

[pswai]

I think it works!

About this one idid rename the the line 2 from dumpy to numpy. and have no idea what im doing just saw the code saying numpy so I did renamed it and it works!? lol. im not sure if this solves all my problems but that was really helpful !😆 God Blessed You By The Computer Gods!!!

File "/Users/lei/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/openpose/body.py", line 2, in import dumpy as np ModuleNotFoundError: No module named 'dumpy'

Glad that it works! For dumpy/numpy it's very weird. For my file it's already numpy. But great that you figured it out! Now have fun!

[shelly-luo]

I got more errors after following this step😭
Startup time: 13.6s (import torch: 0.8s, import gradio: 3.7s, import ldm: 0.3s, other imports: 0.6s, list extensions: 0.3s, load scripts: 0.8s, load SD checkpoint: 6.5s, create ui: 0.2s, gradio launch: 0.3s).
Traceback (most recent call last):
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/gradio/routes.py", line 337, in run_predict
output = await app.get_blocks().process_api(
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/gradio/blocks.py", line 1015, in process_api
result = await self.call_function(
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/gradio/blocks.py", line 833, in call_function
prediction = await anyio.to_thread.run_sync(
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/anyio/to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 867, in run
result = context.run(func, *args)
File "/Users/shelly/Documents/stable-diffusion-webui/extensions/sd-webui-controlnet/scripts/controlnet.py", line 373, in run_annotator
result, is_image = preprocessor(img, res=pres, thr_a=pthr_a, thr_b=pthr_b)
File "/Users/shelly/Documents/stable-diffusion-webui/extensions/sd-webui-controlnet/scripts/processor.py", line 138, in openpose
result = model_openpose(img)
File "/Users/shelly/Documents/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/openpose/init.py", line 84, in call
candidate, subset = self.body_estimation(oriImg)
File "/Users/shelly/Documents/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/openpose/body.py", line 48, in call
Mconv7_stage6_L1, Mconv7_stage6_L2 = self.model(data)
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/shelly/Documents/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/openpose/model.py", line 116, in forward
out1 = self.model0(x)
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/nn/modules/container.py", line 217, in forward
input = module(input)
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/shelly/Documents/stable-diffusion-webui/extensions-builtin/Lora/lora.py", line 202, in lora_Conv2d_forward
return lora_forward(self, input, torch.nn.Conv2d_forward_before_lora(self, input))
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/nn/modules/conv.py", line 463, in forward
return self._conv_forward(input, self.weight, self.bias)
File "/Users/shelly/Documents/stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/nn/modules/conv.py", line 459, in _conv_forward
return F.conv2d(input, weight, bias, self.stride,
RuntimeError: slow_conv2d_forward_mps: input(device='cpu') and weight(device=mps:0') must be on the same device

[sonluong1987]

@shelly-luo I believe u need to add this code to body.py line 48

if torch.backends.mps.is_available():
    mps_device = torch.device("mps:0")
    data = data.to(mps_device)

[AlanZhou2022]

Hi, my problem is that controlnet doesn't take effect when processing pictures in my M2 macbook and no mistakes was showed. (i'm sure that I strictly follow the guidelines, stable diffusion V1.5.1 with the newest version of controlnet) I'm not sure if anybody have the same problem with me?

[JohanBinst]

I'm having the same issue on a M2. I used a xy plot with different control net models and weights and all the output images we're the same.

[foxstudio]

That's great to hear! I'm glad you were able to solve the issue. I hope your experience with the Mac Studio M1 Ultra chip has been helpful to everyone. If you have any other questions or need further assistance, feel free to ask.

File: ../comfyui/custom_node/facerestore/facelib/detection/retinaface/retinaface.py

[Lichtfabrik]

Hi there, followed the instruction to fix this issue, but I still got this Error. Only when I want to use AnimateDiff in combination with ControllNet on A1111.

RuntimeError: slow_conv2d_forward_mps: input(device='cpu') and weight(device=mps:0') must be on the same device

I have added the lines in the body.py etc. Any idea what I could do else?

[AlanZhou2022]

Many errors occurred in Mac, I think SD is not compatible with Mac very well, but it works well in laptop with Nvidia.

…

----------------

----------原邮件信息---------- 发件人："JohanBinst" ***@***.***> 收件人："Mikubill/sd-webui-controlnet" ***@***.***> 抄送人："Alan Zhou" ***@***.***>, "Comment" ***@***.***> 主题：Re: [Mikubill/sd-webui-controlnet] Can't get it to run on Applesilicon. (Discussion #860) 发送时间：2024-02-27 14:48:47 I'm having the same issue on a M2. I used a xy plot with different control net models and weights and all the output images we're the same.

-- Reply to this email directly or view it on GitHub: #860 (reply in thread) You are receiving this because you commented. Message ID: ***@***.***>