TMElyralab / MuseV

MuseV: Infinite-length and High Fidelity Virtual Human Video Generation with Visual Conditioned Parallel Denoising
Other
2.1k stars 214 forks source link

no module named "videocreation" #97

Closed wly-ai-bj closed 2 months ago

xzqjack commented 2 months ago

@wly-ai-bj 请问下有详细的命令和log吗?

wly-ai-bj commented 2 months ago

@wly-ai-bj 请问下有详细的命令和log吗?

按照readme里的方法,

python ./scripts/extract_feature/extract_video_emb_with_multi_process.py -task_path ./datasets/webvid/csvs/train_webvid_10M_train_portrait_and_action_w=512_h=320_10.csv -h5py_dir ./datasets/webvid/emb/train/ -video_dir ./datasets/webvid/video/train -target_width 512 -target_height 320 -source webvid --sep , --n_process 1

直接提示no module named "videocreation",没有其它。

而且在代码里可以看到,这部分import有问题。请问这个是python包还是集成的代码?

image

xzqjack commented 2 months ago

@wly-ai-bj 请问下有详细的命令和log吗?

按照readme里的方法,

python ./scripts/extract_feature/extract_video_emb_with_multi_process.py -task_path ./datasets/webvid/csvs/train_webvid_10M_train_portrait_and_action_w=512_h=320_10.csv -h5py_dir ./datasets/webvid/emb/train/ -video_dir ./datasets/webvid/video/train -target_width 512 -target_height 320 -source webvid --sep , --n_process 1

直接提示no module named "videocreation",没有其它。

而且在代码里可以看到,这部分import有问题。请问这个是python包还是集成的代码?

image

看起来是我们没清理干净,这个是原版的包名字。需要全部改成musev。我们后面更新修复下

xzqjack commented 2 months ago

@wly-ai-bj 已在最新的train分支更新