Zhao (Dylan) Wang

Training Error assert (boxes1[:, 2:] >= boxes1[:, :2]).all()

2

## Instructions To Reproduce the 🐛 Bug: 1. what changes you made (`git diff`) or what code you wrote ``` Nothing change ``` 2. what exact command you run: ```python...

Why the LogME is implemented not in torch but in numpy?

3

RT. Maybe with CUDA, LogME costs less time? Or if this is the problem setting assuming the transferability evaluation can not access GPU? (but the feature extraction part needs GPU?)

Thanks!

Dear Dev, For example, I want the first part show to subsubsection while the second part show to chapter level. Can you help me with this? Looking forward to your...

How to inference for video editing?

5

Hi, Thanks for sharing the great work! I woul dlike to learn how to try video editing after training?

Questions regarding the details of DragonDiffusion

2

Thanks for the great work! I am wondering how you obtain the mask of a specific object?

Is there any suggestions to acccelerate the process of splitting and captioning?

3

RT. As there are a large number of videos for processing, if there is an acceleration, such as multi-processing?

ValueError: Attempting to unscale FP16 gradients.

2

Get an error during stage 2. ``` Traceback (most recent call last): File "/mnt/petrelfs/wangzhao/HumanT2V/AnimateAnyone/train_svd.py", line 803, in main(config) File "/mnt/petrelfs/wangzhao/HumanT2V/AnimateAnyone/train_svd.py", line 672, in main accelerator.clip_grad_norm_( File "/mnt/petrelfs/wangzhao/anaconda3/envs/consisti2v/lib/python3.10/site-packages/accelerate/accelerator.py", line 1925, in...