A short term predicting sequences system

A electronical alert system shows up every 5 minutes, a series of 12 alerts (Each number regarding one alert ,is from a disk error, or a network error, or a program error.. ) . 7 of them , are critical alerts and they are shown first in the file named "results.csv"

the datas are like that way : file results.csv
8,9,11,16,22,27,28
file alerts.csv (first number is the system number, not useful)
199,3,8,9,11,15,16,19,21,22,24,27,28

Those alerts act like a pseudo random number, but, in fact, when an error is solved, the alerts disapeared, but most of the times, they repeat the next row, or the row after. And my goal is to be sure, that errors are correctly solved

I need someone able to utilize the way shown, in this article https://ai.stackexchange.com/questions/3850/can-a-neural-network-be-used-to-predict-the-next-pseudo-random-number"
to predict a next sequence of those alerts.

but, very important, in the code shown, in the link, instead of generating random numbers, to introduce line by line, every sequence of the alerts.csv file

the adaptation of the provided program "example_lstm_read_files.py" needs a python developper