Home

Awesome

Zoo-Tuning

Code release for Zoo-Tuning: Adaptive Transfer from A Zoo of Models (ICML2021)

Pretrained Models

Pretrained ModelsReference
ImageNet Supervisedhttps://pytorch.org/vision/stable/models.html#id10
MoCohttps://github.com/facebookresearch/moco
Mask R-CNNhttps://pytorch.org/vision/stable/models.html#id41
DeepLabV3https://pytorch.org/vision/stable/models.html#deeplabv3
Keypoint R-CNNhttps://pytorch.org/vision/stable/models.html#keypoint-r-cnn

For convenience, we also provide the pretrained models downloaded from these pages. Download

Datasets

DatasetDownload Link
CIFAR-100Downloaded automatically from torchvision.
COCO-70https://github.com/thuml/CoTuning
FGVC Aircrafthttp://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/
Stanford Carshttp://ai.stanford.edu/~jkrause/cars/car_dataset.html
MIT Indoorshttp://web.mit.edu/torralba/www/indoor.html

Requirements

Quick Start

Citation

If you find this code or our paper useful, please consider citing:<br>

@inproceedings{shu2021zoo,
  title={Zoo-Tuning: Adaptive Transfer from a Zoo of Models},
  author={Shu, Yang and Kou, Zhi and Cao, Zhangjie and Wang, Jianmin and Long, Mingsheng},
  booktitle={International Conference on Machine Learning},
  pages={9626--9637},
  year={2021},
  organization={PMLR}
}

Contact

If you have any problems about our code, feel free to contact<br>