search

openai / neural-mmo

Code for the paper "Neural MMO: A Massively Multiagent Game Environment for Training and Evaluating Intelligent Agents"
https://openai.com/blog/neural-mmo/
MIT License
1.57k stars 262 forks source link

Logs don't get updated #32

Closed frenzytejask98 closed 5 years ago

frenzytejask98 commented 5 years ago

The logs folder doesn't update. Moreover how to generate models for 16, 35, 64, etc agents. I am only able to run it for 128 number of agents as I don't have example models for the rest. @jsuarez5341 please help.

jsuarez5341 commented 5 years ago

Hi! If you check the configs.py and experiments.py, you'll notice that the default setup is running in test mode to show off the demo model. The demo was trained with 128 agents -- experiments.py has short (though admittedly messy) example code for setting training parameters, including population size. Logs and models are written to resource/exps (I forgot to remove the logs directory from a previous version and nobody has mentioned it until now). Feel free to reopen if you are still having issues

Also, it's worth mentioning we're only a day or two out from the first major patch. Ideally it will help clear up most of the confusion. I'll add better config docs to my todos for today.