tanu20/unit7-soccer-agent

SoccerTwos MA-POCA Agent

This model was trained for the Hugging Face Deep Reinforcement Learning Course Unit 7 using the ML-Agents SoccerTwos environment and the MA-POCA trainer.

Training configuration:

• Algorithm: MA-POCA
• Environment: SoccerTwos
• Maximum training steps: 5,000,000
• Actual training stopped after approximately 510,000 steps
• Framework: Unity ML-Agents 0.30.0

The repository contains the trained ONNX policy checkpoint.