Questions for the Contest: Pre trained models

[Florian Pirchner]

Hi,

I got some questions for the contest.

• Is it allowed to use pre-trained models in the docker package? Eg pre trained models trained on the labels and depth buffer to support the "reinforcement learning" step?
  
• Is it allowed to use images about the monsters, the life packs, ammo,... to train the model more easily.

Or is the contest ONLY running on the raw pixel data, without any external models and knowledge?

Thanks for your help

Florian Pirchner

[Wojciech Jaskowski]

Florian,

The contest is over, see the results: http://vizdoom.cs.put.edu.pl/competition-cig-2017/results

Best, Wojciech

--
Wojciech Jaśkowski, Ph.D.
Laboratory of Intelligent Decision Support Systems
Institute of Computing Science
Poznan University of Technology, Poland
http://www.cs.put.poznan.pl/wjaskowski

--
You received this message because you are subscribed to the Google Groups "Visual Doom AI Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vizdoom+unsubscribe@googlegroups.com.
To post to this group, send email to viz...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/vizdoom/cc9a8424-cf7b-461a-b80f-cbdd98f04aee%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.