Commit Graph

e947408 (RNN) change hyper parameters by mohamad maheri 2022-03-07 11:07:15 +0330
de53dda Implementing Robustify Negative sampling and undo prepossessing by mohamad maheri 2022-03-06 19:52:06 +0330
dbbec9f repair bug of skipping in sequence by mohamad maheri 2022-02-07 23:47:31 +0330
52f4746 using GRU instead of LSTM by mohamad maheri 2022-01-31 12:25:40 +0330
703b6e5 could skip sequence (p=0.5) by mohamad maheri 2022-01-30 09:11:16 +0330
372e75d hyper parameter tuning including neg_size by mohamad maheri 2021-12-28 15:04:38 +0330
c89c660 hyper parameter tuning including neg_size by mohamad maheri 2021-12-21 17:12:54 +0330
8432f28 implement top1 by mohamad maheri 2021-12-16 18:52:11 +0330
dca86ac implement bpr and bpr_max loss by mohamad maheri 2021-12-16 18:13:37 +0330
703d2aa bpr loss + 5X NS by mohamad maheri 2021-12-14 07:01:05 +0330
6af0db4 Distribute NS to all member of the sequence (except last item of seq in support-set) by mohamad maheri 2021-12-13 04:18:19 +0330
2a680a2 increase NS size (10X) + choose K most scored for NS by mohamad maheri 2021-12-13 01:32:10 +0330
de111c7 negative sampling by choosing from other user's items by mohamad maheri 2021-12-12 20:53:41 +0330
b79f900 make sampler to choose only sequenced items + write results by mohamad maheri 2021-12-07 15:55:49 +0330
8e21c39 cuda setting fixed by mohamad maheri 2021-11-26 19:42:41 +0330
5c17c5e bug fixing for hyper parameter tuning by mohamad maheri 2021-11-15 06:54:58 +0330
6029ca9 using ray for hyper-parameter tanning by mohamad maheri 2021-11-10 06:22:15 +0330
9cbfd2a input all sequence of items to a LSTM (instead of averaging pairs) by mohamad maheri 2021-11-05 02:51:08 +0330
1c015e3 using LSTM instead of MLP by mohamad maheri 2021-11-03 19:06:55 +0330
f3bf760 (HEAD -> master) first commit by mohamad maheri 2021-10-22 13:53:13 +0330
9e1c651 first commit by mohamad maheri 2021-10-22 13:50:17 +0330