CogVideo issues

On the parallel setting

2

I find the codes in arguments.py file below " if args.model_parallel_size

fix: Correct typo in example prompt description

- Changed "it's" to "its" in the example prompt about the SUV on a dirt road. This fixes a grammatical error where "it's" (contraction for "it is") was incorrectly used...

CharlesCNorton

Fail to find "sat.mpu", "sat.model" and "sat.training"

3

from sat.model.base_model import get_model from sat.training.model_io import load_checkpoint from sat import mpu but not find corresponding files

passerby233

RuntimeError: r.nvmlDeviceGetNvLinkRemoteDeviceType_ INTERNAL ASSERT FAILED at "../c10/cuda/driver_api.cpp":27, please report a bug to PyTorch. Can't find nvmlDeviceGetNvLinkRemoteDeviceType: /lib64/libnvidia-ml.so.1: undefined symbol: nvmlDeviceGetNvLinkRemoteDeviceType

1

When I want to use a for loop like this to run multiple results: ```python pipe = CogVideoXPipeline.from_pretrained(...) for prompt in prompts: pipe.enable_sequential_cpu_offload() prompt_embeds, _ = pipe.encode_prompt(prompt=prompt,...) video = pipe(...).frames[0]...

HunYuanfeng

找不到这个diffusers库

1

SCHfighting

position embedding

1

I noticed that position embedding is 3d_sincos_pos_embed, but the position embedding mentioned in the paper is rope. Will the code using rope be released?

nankepan