POCA Multi-Agent playing ML-Agents-SoccerTwos
This is a trained model of multi-agent POCA (Posthumous Credit Assignment) playing ML-Agents-SoccerTwos for Unit 7 of the Hugging Face Deep RL Course.
Evaluation Results
- Mean Reward: 2.0 +/- 0.5
- Score (Mean - Std): 1.5 (Requirement: >= -100.0)
Usage
The model file SoccerTwos.onnx can be embedded into Unity ML-Agents environments.
- Downloads last month
- 8
Evaluation results
- mean_reward on ML-Agents-SoccerTwosself-reported2.0 +/- 0.5