Datasets
You can download datasets directly.
Dataset Name | Classes | Number of Instances (Training set) | Number of Instances (Testing set) | Number of Instances (Validating set) | |
---|---|---|---|---|---|
cifar-10-long-tailPrimary
|
10 | 10415 | 2236 | 2233 | Download |
cifar-100-long-tail
|
100 | 15146 | 3299 | 3249 | Download |
tiny-imagenet-200
|
200 | 70000 | 15000 | 15000 | Download |
imagenet
|
1000 | 115846 | 50000 | 19999 | Download |
Citation
Please cite the corresponding paper if you find this library helpful. For example,
@inproceedings{DBLP:conf/icml/YQBYXQ, author = {Zhiyong Yang, Qianqian Xu, Shilong Bao, Yuan He, Xiaochun Cao and Qingming Huang}, title = {When All We Need is a Piece of the Pie: A Generic Framework for Optimizing Two-way Partial AUC}, booktitle = {ICML}, pages = {11820--11829}, year = {2021} }