unity-tech-cn
/
ml-agents
45
14
派生 39
代码 合并请求 版本发布
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
29 提交
337 分支
128 Plastic标签
 
 
 
 
 
目录树: f60cc019
分支列表 标签列表
${ item.name }
创建分支 ${ searchTerm }
从 'f60cc019'
${ noResults }
ml-agents/ml-agents/mlagents/trainers
GitHub 09c7787c [bug-fix] Fix regression in --initialize-from feature (#4086) 4 年前
..
components [refactor] Structure configuration files into classes (#3936) 4 年前
ghost [refactor] Move checkpoint saving into trainer (#4034) 4 年前
optimizer [refactor] Structure configuration files into classes (#3936) 4 年前
policy [versioning] Save ML-Agents version in checkpoints and check on load (#4035) 4 年前
ppo [cleanup] Remove unused param keys (#4067) 4 年前
sac [refactor] Move checkpoint saving into trainer (#4034) 4 年前
tests [bug-fix] Fix regression in --initialize-from feature (#4086) 4 年前
trainer [cleanup] Remove unused param keys (#4067) 4 年前
__init__.py Increment versions to 0.18.0 and 1.1.0 (#4075) 4 年前
action_info.py Move advance() logic for environment manager out of trainer_controller (#3234) 5 年前
agent_processor.py Renaming max_step to interrupted in TermialStep(s) (#3908) 5 年前
barracuda.py fix errors from new flake8-comprehensions (#2917) 5 年前
behavior_id_utils.py Asymmetric self-play (#3653) 5 年前
brain.py removed extraneous logging imports and loggers 5 年前
brain_conversion_utils.py WIP : Changes to the LL-API - Refactor of “done” logic (#3681) 5 年前
buffer.py Fix clear update buffer when trainer stops training, add test (#3422) 5 年前
cli_utils.py [refactor] Move checkpoint saving into trainer (#4034) 4 年前
curriculum.py [refactor] Structure configuration files into classes (#3936) 4 年前
demo_loader.py Catch dimension mismatches between demos and policy (#3821) 5 年前
distributions.py Hotfixes for Release 0.15.1 (#3698) 5 年前
env_manager.py [WIP] Side Channel Design Changes (#3807) 5 年前
exception.py Combined model and policy for PPO 5 年前
learn.py [bug-fix] Fix regression in --initialize-from feature (#4086) 4 年前
meta_curriculum.py [refactor] Store and restore state along with checkpoints (#4025) 4 年前
models.py Use LR schedule for beta and epsilon (#3940) 5 年前
run_experiment.py [refactor] Structure configuration files into classes (#3936) 4 年前
sampler_class.py Moving Env Manager to Trainers (#3062) The Env Manager is only used by the trainer codebase. The entry point to interact with an environment is UnityEnvironment. 5 年前
settings.py [refactor] Move checkpoint saving into trainer (#4034) 4 年前
simple_env_manager.py Replaced get_behavior_names and get_behavior_spec with behavior_specs property (#3946) 5 年前
stats.py Asymmetric self-play (#3653) 5 年前
subprocess_env_manager.py Update precommit flake8 (#3961) 5 年前
tensorflow_to_barracuda.py backport tf2bc changes from barracuda-release (#3341) 5 年前
trainer_controller.py [refactor] Move checkpoint saving into trainer (#4034) 4 年前
trainer_util.py [refactor] Structure configuration files into classes (#3936) 4 年前
training_status.py [refactor] Store and restore state along with checkpoints (#4025) 4 年前
trajectory.py Renaming max_step to interrupted in TermialStep(s) (#3908) 5 年前
upgrade_config.py [refactor] Improve config upgrade script and add test (#4056) 4 年前