search

jjihwan / FIFO-Diffusion_public

Official implementation of FIFO-Diffusion: Generating Infinite Videos from Text without Training (NeurIPS 2024)
https://jjihwan.github.io
397 stars 26 forks source link

No Module called 'opensora.model.xxxxx' #28

Closed ZhangT-tech closed 1 week ago

ZhangT-tech commented 3 months ago

Hi, thanks for your great work, I m trying with the open-sora-plan code following the Readme doc, but after installing everything, when i run the script opensora_fifo_65.sh, it shows me unable to load the module that is in Open-Sora-Plan.

The directory now is: FIFO-Diffusion-public / --- Open-Sora-Plan / ------opensora / ---------model / ...............

jjihwan commented 3 months ago

Sry, i couldn't understand the issue what is the exact error message ?

ZhangT-tech commented 3 months ago

Oh, thank you so much for your reply, the error message is like this:

(.sora) [zhant0g@gpu609-07 FIFO-Diffusion_public]$ sh ./scripts/opensora_fifo_65.sh ** /ibex/user/zhant0g/code/FIFO-Diffusion_public/Open-Sora-Plan ** /ibex/user/zhant0g/code/FIFO-Diffusion_public/.sora/lib64/python3.9/site-packages/torchvision/transforms/_functional_video.py:6: UserWarning: The 'torchvision.transforms._functional_video' module is deprecated since 0.12 and will be removed in the future. Please use the 'torchvision.transforms.functional' module instead. warnings.warn( /ibex/user/zhant0g/code/FIFO-Diffusion_public/.sora/lib64/python3.9/site-packages/torchvision/transforms/_transforms_video.py:22: UserWarning: The 'torchvision.transforms._transforms_video' module is deprecated since 0.12 and will be removed in the future. Please use the 'torchvision.transforms' module instead. warnings.warn( [2024-08-14 13:38:40,211] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) Warning: The default cache directory for DeepSpeed Triton autotune, /home/zhant0g/.triton/autotune, appears to be on an NFS system. While this is generally acceptable, if you experience slowdowns or hanging when DeepSpeed exits, it is recommended to set the TRITON_CACHE_DIR environment variable to a non-NFS path. [WARNING] async_io requires the dev libaio .so object and headers but these were not found. [WARNING] If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. [WARNING] Please specify the CUTLASS repo directory as environment variable $CUTLASS_PATH [WARNING] NVIDIA Inference is only supported on Ampere and newer architectures [WARNING] sparse_attn requires a torch version >= 1.5 and < 2.0 but detected 2.1 [WARNING] using untested triton version (2.1.0), only 1.0.0 is known to be compatible /ibex/user/zhant0g/code/FIFO-Diffusion_public/.sora/lib64/python3.9/site-packages/torchvision/transforms/functional_tensor.py:5: UserWarning: The torchvision.transforms.functional_tensor module is deprecated in 0.15 and will be removed in 0.17. Please don't rely on it. You probably just need to use APIs in torchvision.transforms.functional or in torchvision.transforms.v2.functional. warnings.warn( /ibex/user/zhant0g/code/FIFO-Diffusion_public/.sora/lib64/python3.9/site-packages/diffusers/models/transformer_2d.py:20: FutureWarning: Transformer2DModelOutput is deprecated and will be removed in version 0.29. Importing Transformer2DModelOutput from diffusers.models.transformer_2d is deprecated and this will be removed in a future version. Please use from diffusers.models.transformers.transformer_2d import Transformer2DModelOutput, instead. deprecate("Transformer2DModelOutput", "0.29", deprecation_message) /ibex/user/zhant0g/code/FIFO-Diffusion_public/.sora/lib64/python3.9/site-packages/diffusers/models/transformer_2d.py:25: FutureWarning: Transformer2DModel is deprecated and will be removed in version 0.29. Importing Transformer2DModel from diffusers.models.transformer_2d is deprecated and this will be removed in a future version. Please use from diffusers.models.transformers.transformer_2d import Transformer2DModel, instead. deprecate("Transformer2DModel", "0.29", deprecation_message) Warning, cannot find cuda-compiled version of RoPE2D, using a slow pytorch version instead Warning, cannot find cuda-compiled version of RoPE2D, using a slow pytorch version instead Traceback (most recent call last): File "/ibex/user/zhant0g/code/FIFO-Diffusion_public/opensora_fifo/sample/sample_fifo.py", line 29, in from opensora.models.ae import ae_stride_config, getae, getae_wrapper ModuleNotFoundError: No module named 'opensora.models.ae'

I later I figure it out it's because I used the V1.2.0, rather than V1.1.0, which causes the module not found issue.