gbildson / nprizeadditions Goto Github PK

Hello Greg,

I just had a look at your C-RBM Code and was wondering if it ever yielded better results than the Basic RBM ? I tried myself an home-made version as close as possible to what is described in all papers (ie Toronto) but it just -NEVER- yielded any better result whatever the parameters were. Despite the fact that the logic behind it seems more than valid I'm starting to think that there is something about that conditional RBM version for Netflix. A misprint ? What do you think ?

Regards.
Keep

Hi! I'm looking at your rbm.c code and I feel like something's strange. Are the ratings used in the program in the range 0-4? The original netflix data was in the scale 1-5, right?

Hey, i am doing a work of optimizing the RMSE of Collaborative Filtering ( i think RBM could perform better combined with louvain, the community dectection algorithm ). And i am searching the qualifying.txt which was said to be provided officially when the competition is over. Would you please send one copy if you still hold one ? 3Q very much

line 225 in rbm.c should be fixed from visbiases[j][i] = 0.02 * randn() - 0.004; to visbiases[j][i] = log( ((double)moviercount[j*SOFTMAX+i]) / ((double) mtot) );?