search

changlin31 / DNA

(CVPR 2020) Block-wisely Supervised Neural Architecture Search with Knowledge Distillation
234 stars 35 forks source link

请问cifar-10迁移学习的那个模型，是直接使用DNA-c模型迁移的吗？ #2

Closed 5663015 closed 4 years ago

5663015 commented 4 years ago

DNA-c模型有5次下采样，cifar-10最后会被压缩为1x1大小的，请问是这样做的吗，谢谢！

jiefengpeng commented 4 years ago

我们采用与谷歌EfficientNet一致的迁移方法，将CIFAR10 resize到(224, 224)再输入网络，所以5次下采样之后是(7, 7). 详情可参考： Kornblith, S., Shlens, J., and Le, Q. V. Do better imagenet models transfer better? CVPR, 2019.