search

jiangtaoxie / fast-MPN-COV

@CVPR2018: Efficient unrolling iterative matrix square-root normalized ConvNets, implemented by PyTorch (and code of B-CNN,Compact bilinear pooling etc.) for training from scratch & finetuning.
http://peihuali.org/iSQRT-COV/index.html
MIT License
270 stars 56 forks source link

Should we use SVMs for FGVC? #12

Closed CupidJay closed 5 years ago

CupidJay commented 5 years ago

Hi, I am reading about your iSQRT paper and i think it is quite interesting. However, I am confused about the usage of SVMs. You wrote "After finetuning, the outputs of iSQRT-COV layer are ℓ2−normalized before inputted to train k one-vs-all linear SVMs with hyperparameter C = 1" in your paper but i didn't find it in your released code.

jiangtaoxie commented 5 years ago

@CupidJay We perform SVM classifier for fair comparison with existing methods. In fact, it improve performance very marginal (<=0.1%). So SVM is not a required option.

CupidJay commented 5 years ago

Thanks for your kind reply.