search

Unity-Technologies / ml-agents

The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.
https://unity.com/products/machine-learning-agents
Other
17.19k stars 4.16k forks source link

CustomObservation is not reset when sending info to brain #2411

Closed oxeg closed 4 years ago

oxeg commented 5 years ago

Unlike vector, visual and text observations, that are reset on every step (see MLAgents.Agent.SendInfoToBrain() method), custom observation is not cleared or reset.

Not sure if it's actually a bug or a feature, but this behaviour feels unnatural.

Environment:

ervteng commented 5 years ago

You're right, this seems like an inconsistent behavior. I've logged the issue and we'll take a look. Thanks!

ervteng commented 4 years ago

Custom observations have been removed in the latest ML-Agents - if you'd like similar behavior, we suggest using Side Channels (https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Python-API.md). Closing this issue for now.

github-actions[bot] commented 3 years ago

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.