Home

Awesome

<div align="center"> <h1>Neural Prompt Search</h1> <div> <a href='https://zhangyuanhan-ai.github.io/' target='_blank'>Yuanhan Zhang</a>&emsp; <a href='https://kaiyangzhou.github.io/' target='_blank'>Kaiyang Zhou</a>&emsp; <a href='https://liuziwei7.github.io/' target='_blank'>Ziwei Liu</a> </div> <div> S-Lab, Nanyang Technological University </div> <img src="figures/motivation.png"> <h3>TL;DR</h3>

The idea is simple: we view existing parameter-efficient tuning modules, including Adapter, LoRA and VPT, as prompt modules and propose to search the optimal configuration via neural architecture search. Our approach is named NOAH (Neural prOmpt seArcH).


<p align="center"> <a href="https://arxiv.org/abs/2206.04673" target='_blank'>[arXiv]</a> • <a href="https://zhangyuanhan-ai.github.io/NOAH" target='_blank'>[project page]</a> </p> </div>

Updatas

[05/2022] arXiv paper has been released.

Environment Setup

conda create -n NOAH python=3.8
conda activate NOAH
pip install -r requirements.txt

Data Preparation

1. Visual Task Adaptation Benchmark (VTAB)

cd data/vtab-source
python get_vtab1k.py

2. Few-Shot and Domain Generation

Quick Start For NOAH

We use the VTAB experiments as examples.

1. Downloading the Pre-trained Model

ModelLink
ViT B/16link

2. Supernet Training

sh configs/NOAH/VTAB/supernet/slurm_train_vtab.sh PATH-TO-YOUR-PRETRAINED-MODEL

3. Subnet Search

sh configs/NOAH/VTAB/search/slurm_search_vtab.sh PARAMETERS-LIMITES

4. Subnet Retraining

sh configs/NOAH/VTAB/subnet/slurm_retrain_vtab.sh PATH-TO-YOUR-PRETRAINED-MODEL

We add the optimal subnet architecture of each dataset in the experiments/NOAH/subnet/VTAB.

5. Performance

fig1

Citation

If you use this code in your research, please kindly cite this work.

@misc{zhang2022neural,
      title={Neural Prompt Search}, 
      author={Yuanhan Zhang and Kaiyang Zhou and Ziwei Liu},
      year={2022},
      eprint={2206.04673},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

Acknoledgments

Part of the code is borrowed from CoOp, AutoFormer, timm and mmcv.

Thanks to Chong Zhou (https://chongzhou96.github.io/) for the code of downloading the VTAB-1k.

<div align="center">

Hits

</div>