Recently I'm confused with semantic hashing. I want to, given a collect of documents, get the semantic hash codes for each document(semantic hashing is introduced here: http://www.cs.toronto.edu/~rsalakhu/papers/semantic_final.pdf). I downloaded the source code of Deep AutoEncoder from http://www.cs.toronto.edu/~hinton/MatlabForSciencePaper.html. It seems helpful but slightly different from semantic hashing especially in the fine-tuning part. That is, I don't know how to get binary codes for each document after fine-tuning. Anyone knows that? Any suggestions on learning semantic hashing are welcomed! Thank you!
[link][6 comments]