fuyuchenIfyw

I modify the training loss function in https://github.com/modelscope/DiffSynth-Studio/blob/8332ecebb70664606e7ff9c4c11feee276744b14/diffsynth/pipelines/wan_video_new.py#L113 ``` def training_loss(self, **inputs): max_timestep_boundary = int(inputs.get("max_timestep_boundary", 1) * self.scheduler.num_train_timesteps) min_timestep_boundary = int(inputs.get("min_timestep_boundary", 0) * self.scheduler.num_train_timesteps) timestep_id = torch.randint(min_timestep_boundary, max_timestep_boundary, (1,)) timestep...

I'm sorry that I ultimately didn't solve the problem. I believe it may have been due to insufficient hardware resources. In the end, I gave up on using MONAI and...