Skip to content

Latest commit

 

History

History
43 lines (34 loc) · 1.97 KB

TEMPLATE.md

File metadata and controls

43 lines (34 loc) · 1.97 KB
Raw

Template for Reproduce configs

Introduction

Name: LibFewShot (template)
Embed.: Conv64F/ResNet12/ResNet18/WRN
Type: Metric/Meta/Fine-tuning
Venue: arXiv'21
Codes: Bold if official, Repo-Name(Italic if unofficial)
Tutorials for train and test this method (if needed)

Cite this work with (template):

@article{li2021LibFewShot,
  title={LibFewShot: A Comprehensive Library for Few-shot Learning},
  author={Li, Wenbin and Dong, Chuanqi and Tian, Pinzhuo and Qin, Tiexin and Yang, Xuesong and Wang, Ziyi and Huo Jing and Shi, Yinghuan and Wang, Lei and Gao, Yang and Luo, Jiebo},
  journal={arXiv preprint arXiv:2109.04898},
  year={2021}
}

Results and Models

Classification

Embedding 📖 miniImageNet (5,1) 💻 miniImageNet (5,1) 📖miniImageNet (5,5) 💻 miniImageNet (5,5) 📝 Comments
1 Conv64F 50.00 ± 0.05 50.00 ± 0.05 ⬇️ 📋 50.00 ± 0.05 50.00 ± 0.05 Comments
2 ResNet121 50.00 ± 0.05 50.00 ± 0.05 50.00 ± 0.05 50.00 ± 0.05 Comments
3 ResNet122 50.00 ± 0.05 50.00 ± 0.05 50.00 ± 0.05 50.00 ± 0.05 Comments

CrossDomain

Embedding mini -> CUB (5,1) mini -> CUB (5,5)
1 Conv64F 50.00 ± 0.05 50.00 ± 0.05
2 ResNet121 50.00 ± 0.05 50.00 ± 0.05
3 ResNet122 50.00 ± 0.05 50.00 ± 0.05

Footnotes

  1. (Ordered footnote) ResNet12-TADAM with [64,128,256,512] ↩ ↩2

  2. (Named footnote) ResNet12-MetaOpt with [64,160,320,640] ↩ ↩2