Daniel Ajisafe

Hi Lucidrains, Thanks for your great work :1st_place_medal: Do we need to keep or re-use this backup [data](https://github.com/lucidrains/se3-transformer-pytorch/tree/main/se3_transformer_pytorch/data)?

Captions

1

How do we update the captions in List of Figures from showing a long text describing the image but only a short title? I tried [this](https://tex.stackexchange.com/questions/2853/tableofcontents-and-listoffigures-help) and [this](https://tex.stackexchange.com/questions/11579/short-captions-for-figures-in-listoffigures) but it...

Cross Attention maps

1

Hello, Thank you so much for your great work and codebase! I would appreciate your clarifications on a few items. 1) From within ```TextToVideoSDPipelineCall.py```, at this [line](https://github.com/hohonu-vicml/TrailBlazer/blob/5389c8bad6413e8b340cefb383c12b78f801c2be/TrailBlazer/Pipeline/TextToVideoSDPipelineCall.py#L283), the attention maps...

Hi, thanks for the great work! I have a setup where I don't have access to the internet on the compute node (where GPU is available) hence can't install using...

XDIT multi GPUs inference code runs OOM with 24 frames

2

@OleehyO thanks for the multi-GPU inference code provided [here](https://github.com/THUDM/CogVideo/blob/main/tools/parallel_inference/parallel_inference_xdit.py) and mentioned [here](https://github.com/THUDM/CogVideo/issues/594#issuecomment-2538497268). But even with two 32GB GPUs, with a similar setup and increasing the number of frames to 24...