Code for training the model

[doron-st]

Dear developers,
I'm trying to understand the code, and repeat your experiments.

It seems from a first glance, that you provided the code for running a trained model on input sequences, outputing prediction.
I can't seem to find the code you used for training the models (creating the weight files), and the input-data file you used for performing the training.

If I'm not mistaken, would it be possible to upload this code to github?

Best regards,
Doron

[Sungtae-Lee]

Dear Doron

Thank you for reaching out to us.

I'm Sungtae Lee who was in charge of deep learning models.
We didn't provide training code of the models since we have not refactored the code as public version.
We are sorry for this inconvenience.
However, the training method we used is nothing special but just standard deep learning training which can be also found in "https://github.com/tensorflow/tensorflow/tree/master/tensorflow/examples/tutorials/mnist"
You can reference our paper to find architectural details that we used. Basically, substituting the model architecture and dataset will do.
Though we are hesitant on providing messy training code, we are happy to help you reproduce the results. Please mail me when you are stuck into some problem or found some missing details while writing the code.

Best Regards,
Sungtae Lee