Statos6
/

poca-SoccerTwos

Reinforcement Learning

deep-reinforcement-learning

ML-Agents-SoccerTwos

Model card Files Files and versions Metrics Training metrics Community

poca-SoccerTwos

1 contributor

History: 3 commits

Statos6's picture

Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better

f5c5d35 verified 10 months ago

SoccerTwos
Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better 10 months ago
SoccerTwos_orig
Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better 10 months ago
run_logs
Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better 10 months ago
.gitattributes

1.52 kB

initial commit 10 months ago
README.md

1.35 kB

Trained with optuna hyperparam optimization with ml-agents-optuna 10 months ago
SoccerTwos.onnx

896 kB
LFS

Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better 10 months ago
config.json

1.97 kB

Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better 10 months ago
configuration.yaml

2.04 kB

Retrained to 10M steps, with higher play_against_latest_model_ratio (0.75 instead of 0.25) this helped the model to learn to play defense better 10 months ago