lateset Diffuser: Inference T2V meets patch_size_t error in CogVideoX-2B/5B

[onevfall]

System Info / 系統信息

Diffuser 0.32.1 does not work but Diffuser 0.31.0 works.

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

1. run the cli_demo.py and load 2B/5B model, and we will get this error. Seems that the latest diffuser version does not consider the 2B/5B compatibility.
2. My detailed error:
   INFO:root:Using default resolution (720, 480) for cogvideox-5b
   Loading pipeline components...: 40%|████████████████████ | 2/5 [00:06<00:08, 2.90s/it]The config attributes {'invert_scale_latents': False} were passed to AutoencoderKLCogVideoX, but are not expected and will be ignored. Please verify your config.json configuration file.
   Loading checkpoint shards: 100%|███████████████████████████████████████████████████████| 2/2 [00:03<00:00, 1.72s/it]
   Loading pipeline components...: 100%|██████████████████████████████████████████████████| 5/5 [00:10<00:00, 2.20s/it]
   Traceback (most recent call last):
   File "/group/40034/minghongcai/VideoMorpher/main.py", line 150, in
   output = pipeline(
   File "/data/miniconda3/envs/myenv/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
   return func(*args, **kwargs)
   File "/group/40034/minghongcai/VideoMorpher/model.py", line 810, in call
   patch_size_t = self.transformer.config.patch_size_t
   AttributeError: 'FrozenDict' object has no attribute 'patch_size_t'. Did you mean: 'patch_size'?

Expected behavior / 期待表现

The latest diffuser version can consider the 2B/5B compatibility.

[OleehyO]

cogvideox-5b-t2v, diffuser 0.32.1 runs normally in my environment, suggest checking if the latest repository has been pulled

[onevfall]

cogvideox-5b-t2v, diffuser 0.32.1 runs normally in my environment, suggest checking if the latest repository has been pulled

@OleehyO Thanks. I seem to have said the wrong version, it was diffusers-0.33.0.dev0. I could not run cogvideox-5b-t2v normally