RuntimeError: shape '[2, 6, 2, 30, 2, 45, 2, 16]' is invalid for input of size 2246400

[NXZXH]

System Info / 系統信息

Help！
hidden_states三三三 torch.Size([2, 13, 16, 60, 90])
encoder_hidden_states三三三 torch.Size([2, 226, 4096])
0%| | 0/50 [00:02<?, ?it/s]
Traceback (most recent call last):
File "/root/miniconda3/envs/CogVideo/lib/python3.10/runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/root/miniconda3/envs/CogVideo/lib/python3.10/runpy.py", line 86, in _run_code
exec(code, run_globals)
File "/root/autodl-tmp/CogVideo/cogvideox-controlnet/inference/cli_demo.py", line 201, in
generate_video(
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/root/autodl-tmp/CogVideo/cogvideox-controlnet/inference/cli_demo.py", line 147, in generate_video
video_generate = pipe(
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/root/autodl-tmp/CogVideo/cogvideox-controlnet/controlnet_pipeline.py", line 625, in call
noise_pred = self.transformer(
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/accelerate/hooks.py", line 170, in new_forward
output = module._old_forward(*args, **kwargs)
File "/root/autodl-tmp/CogVideo/cogvideox-controlnet/cogvideo_transformer.py", line 49, in forward
hidden_states = self.patch_embed(encoder_hidden_states, hidden_states)
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/root/miniconda3/envs/CogVideo/lib/python3.10/site-packages/diffusers/models/embeddings.py", line 724, in forward
image_embeds = image_embeds.reshape(
RuntimeError: shape '[2, 6, 2, 30, 2, 45, 2, 16]' is invalid for input of size 2246400

2. Patch embedding

    print("hidden_states三三三", hidden_states.shape)
    print("encoder_hidden_states三三三", encoder_hidden_states.shape)
    hidden_states = self.patch_embed(encoder_hidden_states, hidden_states)

Information / 问题信息

• The official example scripts / 官方的示例脚本
• My own modified scripts / 我自己修改的脚本和任务

Reproduction / 复现过程

。

Expected behavior / 期待表现

。