fcakir/deep-mihash

Code for papers "Hashing with Mutual Information" (TPAMI 2019) and "Hashing with Binary Matrix Pursuit" (ECCV 2018)

deep-mihash's Issues

Bad results

I run with following comman:

demo_AP('cifar',32,'alexnet_ft','split',1,'nbins',32,'sigmf', [1 0],'lr', 1e-3,'lrstep',50,'epoch',100,'obj','mi','testInterval',10, 'batchSize', 256, 'metrics', 'AP')

However, the final mAP is very low (about 0.55).

Can you share the results on AlexNet and cifar-s1?

binary codes are not distinct

This method produces very limited distinct binary codes!- What is the difference between this method and simple one hot encoding?