Prabhjyot's Status Report #6 of 17

[Prabhjyot Singh Sodhi <prabhjyotsingh95 () gmail com>]

Hey all,

I started with fixing the problem with the opencv implementation of the
random forest, following which I moved on to changing database
representation. I was sick for a couple of days and wasn't able to cover a
lot in those days.

Achievements -
- Fixed opencv implementation. Opencv was considering the problem to be a
regression problem and hence was returning averaged out values instead of
class numbers.
- Identified the types of cases that are to be handled by for the one group
per class representation.

Priorities -
- Change the sklearn implementation to return top three closest classes
with accuracies.
- form a general idea about the working of the multi-stage classifier
- Change the database representation to the one group per class and test
the random forest models on that db.

Cheers,
Prabhjyot

_______________________________________________
Sent through the dev mailing list
https://nmap.org/mailman/listinfo/dev
Archived at http://seclists.org/nmap-dev/