Unity 机器学习代理工具包 (ML-Agents) 是一个开源项目,它使游戏和模拟能够作为训练智能代理的环境。
您最多选择25个主题 主题必须以中文或者字母或数字开头,可以包含连字符 (-),并且长度不得超过35个字符
 
 
 
 
 
vincentpierre 5d384292 forgot one 4 年前
..
ghost [🐛 🔨 ]Adding the ELO to the GlobalTrainingStatus (#5202) 4 年前
model_saver [bug-fix] Delete .pt checkpoints past keep-checkpoints (#5271) 4 年前
optimizer [bug-fix] Fix POCA LSTM, pad sequences in the back (#5206) 4 年前
poca Removing misleading learning rate 4 年前
policy [bug-fix] Delete .pt checkpoints past keep-checkpoints (#5271) 4 年前
ppo forgot one 4 年前
sac Removing misleading learning rate 4 年前
tests Fix the reporting of histogram stats and adding a test (#5410) 4 年前
torch Removing misleading learning rate 4 年前
trainer [debug] Require all behavior names to have a matching YAML entry (#5210) 4 年前
__init__.py Version bump main (#5279) 4 年前
action_info.py Move the Critic into the Optimizer (#4939) 4 年前
agent_processor.py Fix the reporting of histogram stats and adding a test (#5410) 4 年前
behavior_id_utils.py Python Dataflow for Group Manager (#4926) 4 年前
buffer.py [bug-fix] Fix POCA LSTM, pad sequences in the back (#5206) 4 年前
cli_utils.py Fix --results-dir (#5269) 4 年前
demo_loader.py Buffer key enums (#4907) 4 年前
directory_utils.py [life improvement] Moving Python files around (#4531) 4 年前
env_manager.py [MLA-1540] Training Analytics (#4780) 4 年前
environment_parameter_manager.py Logging lesson number at the begining of training (#5199) 4 年前
exception.py [add-fire] Merge post-0.19.0 master into add-fire (#4328) 4 年前
learn.py Turning some logger.info into logger.debug and remove some logging overhead when not using debug (#5211) 4 年前
run_experiment.py [refactor] Structure configuration files into classes (#3936) 5 年前
settings.py Adding a fully connected visual encoder for super small visual input + tests (#5351) 4 年前
simple_env_manager.py Merge master into hybrid actions staging branch (#4704) 4 年前
stats.py collecting latest step as a stat (#5264) 4 年前
subprocess_env_manager.py Turning some logger.info into logger.debug and remove some logging overhead when not using debug (#5211) 4 年前
trainer_controller.py Turning some logger.info into logger.debug and remove some logging overhead when not using debug (#5211) 4 年前
training_status.py [bug-fix] Delete .pt checkpoints past keep-checkpoints (#5271) 4 年前
trajectory.py [perf] Optimizations for performance (#5192) 4 年前
upgrade_config.py Update add-fire to latest master, including Policy refactor (#4263) 4 年前