UniAdapter

[ICLR2024] The official implementation of paper "UniAdapter: Unified Parameter-Efficient Transfer Learning for Cross-modal Modeling", by Haoyu Lu, Yuqi Huo, Guoxing Yang, Zhiwu Lu, Wei Zhan, Masayoshi Tomizuka, Mingyu Ding.

Getting Started

• Python3, PyTorch>=1.8.0, torchvision>=0.7.0 are required for the current codebase.
• To install the other dependencies, run

  pip install -r requirements.txt

Image-text Retrieval

• Download COCO and Flickr30k datasets from the original websites, and set 'image_root' in configs/retrieval_{dataset}.yaml accordingly.

• To parameter-efficient finetune on MSCOCO/Flickr:

python -m torch.distributed.run --nproc_per_node=8 train_retrieval.py --config ./configs/retrieval_{coco, flickr}.yaml --output_dir output/{coco, flickr} 

• To evaluate UniAdapter on MSCOCO/Flickr:

python -m torch.distributed.run --nproc_per_node=8 train_retrieval.py --config ./configs/retrieval_{coco, flickr}.yaml --output_dir output/{coco, flickr} --evaluate 

Visual Question Answerring

• Download VQA v2 dataset and Visual Genome dataset from the original websites, and set 'vqa_root' and 'vg_root' in configs/vqa.yaml.

• To parameter-efficient finetune on VQAv2:

python -m torch.distributed.run --nproc_per_node=8 train_vqa.py --config ./configs/vqa.yaml --output_dir $static_dir

• To evaluate UniAdapter on VQAv2 (need to update the result file to the official server):

python -m torch.distributed.run --nproc_per_node=8 train_vqa.py --config ./configs/vqa.yaml --output_dir $static_dir --evaluate 

Video-text Retrieval and VideoQA

• In progress.

Acknowledgement

Our codebase is built based on BLIP, timm. We thank the authors for the nicely organized code!