TypeError: WanS2VModel.patchify() takes 2 positional arguments but 3 were given

pipe(
    prompt=prompt,
    input_image=input_image,
    negative_prompt=negative_prompt,
    seed=0,
    num_frames=num_frames,
    height=height,
    width=width,
    audio_sample_rate=sample_rate,
    input_audio=input_audio,
    num_inference_steps=40,
    sliding_window_size=48, ### 去掉这两处就不会有报错
    sliding_window_stride=24, ### 
)

使用 s2v 模型推理时，如果添加 sliding_window_size 和 liding_window_stride 这两个参数，会有报错

Traceback (most recent call last):
  File "/.projector/apps/pycharm-community-2021.3.2/plugins/python-ce/helpers/pydev/_pydevd_bundle/pydevd_exec2.py", line 3, in Exec
    exec(exp, global_vars, local_vars)
  File "<input>", line 1, in <module>
  File "/llms/DiffSynth-Studio/diffsynth/pipelines/wan_video_new.py", line 1413, in model_fn_wan_video
    return TemporalTiler_BCTHW().run(
  File "/DiffSynth-Studio/diffsynth/pipelines/wan_video_new.py", line 1346, in run
    model_output = model_fn(**model_kwargs).to(device=data_device, dtype=data_dtype)
  File "/DiffSynth-Studio/diffsynth/pipelines/wan_video_new.py", line 1491, in model_fn_wan_video
    x = dit.patchify(x, control_camera_latents_input)
TypeError: WanS2VModel.patchify() takes 2 positional arguments but 3 were given

似乎是因为在 /llms/DiffSynth-Studio/diffsynth/pipelines/wan_video_new.py 判断然后走了这个 TemporalTiler_BCTHW().run 分支而不是后面的分支导致的

Start in diffsynth/pipelines/wan_video_new.py around lines 1346, 1413, and 1491, then inspect the WanS2VModel.patchify signature and the TemporalTiler_BCTHW().run branch. Reproduce the failure with sliding_window_size=48 and sliding_window_stride=24; done means S2V inference accepts those arguments without the reported TypeError.