Sam2Segmentation mps runtime error Input type (float) and bias type (c10::Half) should be the same

[orbenda83]

Hello,

I am running comfyui on M1 macbook and when the sam2segmentation node starts it gets the following error:
!!! Exception during processing !!! Input type (float) and bias type (c10::Half) should be the same

I have set the Load sam model with mps and pf16.

File "/Users/orbend/pinokio/api/comfyui/app/execution.py", line 323, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "/Users/orbend/pinokio/api/comfyui/app/execution.py", line 198, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "/Users/orbend/pinokio/api/comfyui/app/execution.py", line 169, in _map_node_over_list
process_inputs(input_dict, i)
File "/Users/orbend/pinokio/api/comfyui/app/execution.py", line 158, in process_inputs
results.append(getattr(obj, func)(**inputs))
File "/Users/orbend/pinokio/api/comfyui/app/custom_nodes/ComfyUI-segment-anything-2/nodes.py", line 314, in segment
model.set_image(image_np[i])
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/custom_nodes/ComfyUI-segment-anything-2/sam2/sam2_image_predictor.py", line 97, in set_image
backbone_out = self.model.forward_image(input_image)
File "/Users/orbend/pinokio/api/comfyui/app/custom_nodes/ComfyUI-segment-anything-2/sam2/modeling/sam2_base.py", line 469, in forward_image
backbone_out = self.image_encoder(img_batch)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/custom_nodes/ComfyUI-segment-anything-2/sam2/modeling/backbones/image_encoder.py", line 31, in forward
features, pos = self.neck(self.trunk(sample))
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/custom_nodes/ComfyUI-segment-anything-2/sam2/modeling/backbones/hieradet.py", line 283, in forward
x = self.patch_embed(x)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/custom_nodes/ComfyUI-segment-anything-2/sam2/modeling/backbones/utils.py", line 92, in forward
x = self.proj(x)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/conv.py", line 460, in forward
return self._conv_forward(input, self.weight, self.bias)
File "/Users/orbend/pinokio/api/comfyui/app/env/lib/python3.10/site-packages/torch/nn/modules/conv.py", line 456, in _conv_forward
return F.conv2d(input, weight, bias, self.stride,
RuntimeError: Input type (float) and bias type (c10::Half) should be the same

Please help on this.

[balmasi]

Seeing the same thing on M2, even on CPU and fp32