devraj89/LABEL-CONSISTENT-MATRIX-FACTORIZATION-BASED-HASHING-FOR-CROSS-MODAL-RETRIEVAL

wiki_ours2

zyfsa opened this issue · 2 comments

zyfsa commented

hi，I have a question about the sampleInds = sampleInds(1:N). where I can find it. Now I want to train my own dataset

devraj89 commented

Hi

Thanks for your interest in our work. So basically the sampleInds was used to select a random number of samples to constitute the training dataset. As far as I remember for MirFlickr and NUS-Wide sampleInds was a vector of 5000 numbers. So when we did sampleInds = sampleInds(1:N) we were just selecting the data for getting the training set.

For your own dataset suppose you want to select N number of samples for your training
simply do this
sampleInds = 1:N
I hope this clears your doubt. Please get in touch with me in case of any further issues.

152ls12 commented

Hi
error:Undefined function 'sampleInds' for input arguments of type 'double',what should I do.