about reproduce the performance about gamenet HOT 2 OPEN

Thanks for pointing out the problem!
This problem is caused by the mismatch between the training data in the Github repository and the training data in our paper.

Way to fix this problem.

If you re-generate the training data using EDA.ipynb, it will provide you with the same data in our paper.

Why higher performance?

The data in the GitHub repository may also be generated by uncommenting a line in EDA.ipynb:

#med_pd = filter_first24hour_med(med_pd)

It means that the medications for a visit are collected beyond 1 day. Thus, the performance may be improved due to more data provided for a patient.
But as we have claimed in our paper, the first 24-hour is often the most critical time for patients to obtain correct treatment quickly. However, it depends on you to choose which version of data in your work.

from gamenet.

I tried running EDA.ipynb but getting this error 'ndc_rxnorm_file' is not defined. Upon further investigation I figured out there is a mismatch in file name. Please note that.

from gamenet.