weipingpku

Thanks Xiang for your clarification. Does such evaluation procedure have suspicion of overfitting the test data? Because part (even half) of the test data is seen during the training.

Thanks for your help! Then there may be some information leakage when appling `transform()` on time-series data.

If I remember correctly, you are right that two session have the same time_id only if their date and timestamps are exactly the same. Please let me if you still...

Thanks for discussing this! You are right this is the implement in the original Hidash's code. While I agree that using tanh as final activation function maybe problematic, especially together...

Hi @IcyLiGit , Sorry about that you can't reproduce the results. First, I didn't test the code with "top1" loss, thus I have no idea how to set "good" parameters,...

It seems that TF converges faster than Theano. So how about: 1. decease training steps. 2. more concretely, use validation data to prevent over-fitting(early stop). Weiping

Hi @sunxiaobiu @hunkim, I am confused with initial_state, too. After a batch, you feed `feed[c]=state[i].c, feed[h]=state[i].h` to model. Does this mean using last state of last batch as initial state...

Thanks for your prompt reply. I'm wondering is it possible to make it compatible with torch.jit.trace for deployment need?