Skip to content

trying to get rife to work with multigpu #105

@clsferguson

Description

@clsferguson
Image Image

I keep getting this error after the vae decodes and rife kicks in. Not really sure how to proceed with this.

!!! Exception during processing !!! Expected all tensors to be on the same device, but got weight is on cuda:4, different from other tensors on cuda:0 (when checking argument in method wrapper_CUDA__cudnn_convolution)

Traceback (most recent call last):

  File "/app/ComfyUI/execution.py", line 496, in execute

    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)

                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/app/ComfyUI/execution.py", line 315, in get_output_data

    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)

                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/app/ComfyUI/execution.py", line 289, in _async_map_node_over_list

    await process_inputs(input_dict, i)

  File "/app/ComfyUI/execution.py", line 277, in process_inputs

    result = f(**inputs)

             ^^^^^^^^^^^

  File "/app/ComfyUI/custom_nodes/comfyui-frame-interpolation/vfi_models/rife/__init__.py", line 104, in vfi

    generic_frame_loop(type(self).__name__, frames, clear_cache_after_n_frames, multiplier, return_middle_frame, *args,

  File "/app/ComfyUI/custom_nodes/comfyui-frame-interpolation/vfi_utils.py", line 222, in generic_frame_loop

    return _generic_frame_loop(

           ^^^^^^^^^^^^^^^^^^^^

  File "/app/ComfyUI/custom_nodes/comfyui-frame-interpolation/vfi_utils.py", line 170, in _generic_frame_loop

    middle_frame = return_middle_frame_function(

                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/app/ComfyUI/custom_nodes/comfyui-frame-interpolation/vfi_models/rife/__init__.py", line 98, in return_middle_frame

    return model(frame_0, frame_1, timestep, scale_list, in_fast_mode, in_ensemble)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl

    return self._call_impl(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl

    return forward_call(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/app/ComfyUI/custom_nodes/comfyui-frame-interpolation/vfi_models/rife/rife_arch.py", line 406, in forward

    f0 = self.encode(img0[:, :3])

         ^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl

    return self._call_impl(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl

    return forward_call(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/container.py", line 244, in forward

    input = module(input)

            ^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl

    return self._call_impl(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl

    return forward_call(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/conv.py", line 548, in forward

    return self._conv_forward(input, self.weight, self.bias)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/usr/local/lib/python3.12/site-packages/torch/nn/modules/conv.py", line 543, in _conv_forward

    return F.conv2d(

           ^^^^^^^^^

RuntimeError: Expected all tensors to be on the same device, but got weight is on cuda:4, different from other tensors on cuda:0 (when checking argument in method wrapper_CUDA__cudnn_convolution)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions