21 Commits (RNN)
 

Author SHA1 Message Date
  mohamad maheri e9474085ef change hyper parameters 2 years ago
  mohamad maheri de53dda45f Implementing Robustify Negative sampling and undo prepossessing 2 years ago
  mohamad maheri dbbec9facc repair bug of skipping in sequence 2 years ago
  mohamad maheri 52f474670f using GRU instead of LSTM 2 years ago
  mohamad maheri 703b6e570b could skip sequence (p=0.5) 2 years ago
  mohamad maheri 372e75db43 hyper parameter tuning including neg_size 2 years ago
  mohamad maheri c89c660a45 hyper parameter tuning including neg_size 2 years ago
  mohamad maheri 8432f281de implement top1 2 years ago
  mohamad maheri dca86acb3a implement bpr and bpr_max loss 2 years ago
  mohamad maheri 703d2aaf76 bpr loss + 5X NS 2 years ago
  mohamad maheri 6af0db43ad Distribute NS to all member of the sequence (except last item of seq in support-set) 2 years ago
  mohamad maheri 2a680a294f increase NS size (10X) + choose K most scored for NS 2 years ago
  mohamad maheri de111c73dc negative sampling by choosing from other user's items 2 years ago
  mohamad maheri b79f900501 make sampler to choose only sequenced items + write results 2 years ago
  mohamad maheri 8e21c390fc cuda setting fixed 2 years ago
  mohamad maheri 5c17c5eb38 bug fixing for hyper parameter tuning 2 years ago
  mohamad maheri 6029ca93a1 using ray for hyper-parameter tanning 2 years ago
  mohamad maheri 9cbfd2aff6 input all sequence of items to a LSTM (instead of averaging pairs) 2 years ago
  mohamad maheri 1c015e3b35 using LSTM instead of MLP 2 years ago
  mohamad maheri f3bf760148 first commit 2 years ago
  mohamad maheri 9e1c651cfe first commit 2 years ago