Image
Image
Image
Image
Image
Image
Image
Image
Image
Image



Search
»

AugARP

Description: The package includes the Python implementation of the method proposed in 1, which aims to augment new instances for cost-aware long-tailed scenario. The proposed method is based on adaptive region partition in the feature space by introducing an attention module into the convolutional network, which can be used in companion with any current instance augmentation approaches.

References:
[1] Yu-Cheng He, Yao-Xiang Ding, Han-Jia Ye and Zhi-Hua Zhou. Learning Only When It Matters: Cost-Aware Long-Tailed Classification. AAAI2024. in press.

ATTN: This package is free for academic usage. You can run it at your own risk. For other purposes, please contact .

ATTN2: This package was developed by . For any problem concerning the code, please feel free to contact Mr. He.

Requirement: The package was developed with Python 3.8.

Download: [code] (49KB)
  Name Size

Image
PoweredBy © LAMDA, 2022