Home

Awesome

<h3><a href="">Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models</a></h3> <a href="https://varybase.github.io/"><img src="https://img.shields.io/badge/Project-Page-Green"></a> <a href="https://arxiv.org/abs/2312.06109"><img src="https://img.shields.io/badge/Paper-PDF-orange"></a> <a href="http://region-31.seetacloud.com:22701/"><img src="https://img.shields.io/badge/demo-blue"></a> <a href="https://zhuanlan.zhihu.com/p/671420712"><img src="https://img.shields.io/badge/zhihu-yellow"></a>

<a href="https://trendshift.io/repositories/5978" target="_blank"><img src="https://trendshift.io/api/badge/repositories/5978" alt="Ucas-HaoranWei%2FVary | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>

Haoran Wei*, Lingyu Kong*, Jinyue Chen, Liang Zhao, Zheng Ge, Jinrong Yang, Jianjian Sun, Chunrui Han, Xiangyu Zhang

<p align="center"> <img src="assets/logo.jpg" style="width: 200px" align=center> </p>

Release

Code License Data License Usage and License Notices: The data, code, and checkpoint are intended and licensed for research use only. They are also restricted to use that follow the license agreement of LLaMA, Vicuna, GPT-4, Qwen, and LLaVA.

Contents

Install

  1. Clone this repository and navigate to the Vary folder
git clone https://github.com/Ucas-HaoranWei/Vary.git
cd Vary
  1. Install Package
conda create -n vary python=3.10 -y
conda activate vary
pip install e .
  1. Install Flash-Attention
pip install ninja
pip install flash-attn --no-build-isolation

Vary Weights

Demo

  1. Update the CLIP-VIT path in the codes (/cache/vit-large-patch14/) to your path.

python vary/demo/run_qwen_vary.py  --model-name  /vary/model/path/ --image-file /an/image/file.png

Train

  1. For Vary-base (one machine, if you have multiple machines you need to prepare your host file)
deepspeed   Vary/train/train_qwen_vary.py  --deepspeed /Vary/zero_config/zero2.json
            --model_name_or_path /Qwen-7B/path/
            --vision_tower /vit-large-patch14/path/
            --freeze_vision_tower True
            --freeze_lm_model False
            --vision_select_layer  -2
            --use_im_start_end True
            --bf16 True
            --per_device_eval_batch_size 4
            --gradient_accumulation_steps 1
            --evaluation_strategy "no"
            --save_strategy "steps"
            --save_steps 5000
            --save_total_limit 1
            --weight_decay 0.
            --warmup_ratio 0.03
            --lr_scheduler_type "cosine"
            --logging_steps 1 --tf32 True
            --model_max_length 4096
            --gradient_checkpointing True
            --dataloader_num_workers 4
            --report_to none
            --per_device_train_batch_size 4
            --num_train_epochs 1
            --learning_rate 5e-5
            --datasets  data_name1+data_name2+data_name3
            --output_dir /path/to/output/
  1. For Vary-tiny
deepspeed   Vary/train/train_opt.py  --deepspeed /Vary/zero_config/zero2.json
            --model_name_or_path /opt125m/path/
            --conversation_version opt
            --freeze_vision_tower False
            --freeze_lm_model False
            --use_im_start_end True
            --bf16 True
            --per_device_eval_batch_size 4
            --gradient_accumulation_steps 1
            --evaluation_strategy "no"
            --save_strategy "steps"
            --save_steps 5000
            --save_total_limit 1
            --weight_decay 0.
            --warmup_ratio 0.03
            --lr_scheduler_type "cosine"
            --logging_steps 1 --tf32 True
            --model_max_length 4096
            --gradient_checkpointing True
            --dataloader_num_workers 4
            --report_to none
            --per_device_train_batch_size 16
            --num_train_epochs 1
            --learning_rate 5e-5
            --datasets  data_name1+data_name2+data_name3
            --output_dir /path/to/output/

Contact

If you have any questions related to the code or the paper, feel free to email (weihaoran18@mails.ucas.ac.cn).

Acknowledgement

Citation

If you find our work useful in your research, please consider citing Vary:

@article{wei2023vary,
  title={Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models},
  author={Wei, Haoran and Kong, Lingyu and Chen, Jinyue and Zhao, Liang and Ge, Zheng and Yang, Jinrong and Sun, Jianjian and Han, Chunrui and Zhang, Xiangyu},
  journal={arXiv preprint arXiv:2312.06109},
  year={2023}
}

@article{wei2024small,
  title={Small Language Model Meets with Reinforced Vision Vocabulary},
  author={Wei, Haoran and Kong, Lingyu and Chen, Jinyue and Zhao, Liang and Ge, Zheng and Yu, En and Sun, Jianjian and Han, Chunrui and Zhang, Xiangyu},
  journal={arXiv preprint arXiv:2401.12503},
  year={2024}
}