search

PathmindAI / nativerl

Train reinforcement learning agents using AnyLogic or Python-based simulations
Apache License 2.0
19 stars 4 forks source link

Add checks for expected metrics levels #500

Closed slinlee closed 2 years ago

slinlee commented 3 years ago

so we can make sure issues like #441 and https://github.com/SkymindIO/nativerl/pull/499 don't regress

this specific test would go in test_training.py, and test a multi agent model like multi mouse and cheese. it might be tough though since the metric is either one or zero