Shilin Xu

The ckpt has a social decoder, but the code does not.

1

Thanks for your code. I encountered several questions when reproducing this paper. I hope you can help me solve these questions. - Code did not have a social decoder, but...

When I convert an object segment into a semantic mask using coco API, I find some intersection areas between the different objects. How should I solve the overleaping area when...

Hi @hongfz16 Thanks for your code. How can I evaluate the LSTQ metric using 4D DS-Net?

Hi @MehmetAygun @jbehley Thank you for sharing your excellent open-source code and the visualization video on your [homepage](https://mehmetaygun.github.io/4DPLS.html). I would greatly appreciate it if you could provide instructions on how...

Can you describe the argoverse data structure in your train/data folder? like command `tree train/data` thanks @GentleSmile

some checkpoints?

2

Would you please provide some checkpoints you trained?

one file to n servers

3

Thank you for this project !!! Can you support one file to n different servers. Because Remote-SSH supports editing files on the server side, the one file to one server...

Hi, Would you please support the pytorch dataloader for the motion dataset? Thanks for your work.

What is the data_velodyne under SemanticKITTI folder

2

What is the data_velodyne under the SemanticKITTI folder? How can I get this?

the loss is nan when pre-training tinyllama using share recipe

1

Here is my training script ``` deepspeed tinyllava/train/train.py \ --deepspeed ./scripts/zero2.json \ --model_name_or_path checkpoints/TinyLlama-1.1B-Chat-v1.0/ \ --version plain \ --data_path datasets/LLaVA-Pretrain/blip_laion_cc_sbu_558k.json \ --image_folder datasets/LLaVA-Pretrain/images \ --vision_tower checkpoints/clip-vit-large-patch14-336 \ --pretrain_mm_mlp_adapter output/pretrain/llava-tinyllama-1.1b/mm_projector.bin \...