At first glance, everything looks ok... I have to ask someone else if there is any problem getting anomaly scores out of a model that is doing category prediction...
--------- Matt Taylor OS Community Flag-Bearer Numenta On Tue, Nov 17, 2015 at 10:47 AM, Sanket Desai <[email protected]> wrote: > Hi Matthew, > > Thank you very much for your reply. I have put my code on github at > https://github.com/snketdesai/NuPic_Network_Anomaly_Detection. I have > been following HotGym tutorial, so you will find my code almost similar to > that one. > I was using, > nupic_output.py - initially for prediction but not for anomaly score. > > nupic_anomaly_output.py file for printing anomaly scores in cvs file, but > it didn’t work out for some reason. So for now, anomaly score is being > printed on console. > > I have uploaded some portion of my cvs data as well if that can help and > will try to include timestamp as per your suggestion. > > Thanks, > Sanket >
