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
 
 
 
 
 

70 行
1.7 KiB

behaviors:
SoccerTwos:
trainer_type: ppo
hyperparameters:
batch_size: 4096
buffer_size: 40960
learning_rate: 0.0003
beta: 0.005
epsilon: 0.2
lambd: 0.95
num_epoch: 3
learning_rate_schedule: constant
network_settings:
normalize: false
hidden_units: 512
num_layers: 3
vis_encode_type: simple
reward_signals:
extrinsic:
gamma: 0.99
strength: 1.0
keep_checkpoints: 5
max_steps: 100000000
time_horizon: 1000
summary_freq: 10000
threaded: false
self_play:
save_steps: 50000
team_change: 200000
swap_steps: 2000
window: 10
play_against_latest_model_ratio: 0.5
initial_elo: 1200.0
environment_parameters:
ball_touch:
curriculum:
- name: Lesson0
completion_criteria:
measure: progress
behavior: SoccerTwos
signal_smoothing: true
min_lesson_length: 100
threshold: 0.001
value: 1.0
- name: Lesson1
completion_criteria:
measure: progress
behavior: SoccerTwos
signal_smoothing: true
min_lesson_length: 100
threshold: 0.00175
value: 0.5
- name: Lesson2
completion_criteria:
measure: progress
behavior: SoccerTwos
signal_smoothing: true
min_lesson_length: 100
threshold: 0.00225
value: 0.1
- name: Lesson3
completion_criteria:
measure: progress
behavior: SoccerTwos
signal_smoothing: true
min_lesson_length: 100
threshold: 0.003
value: 0.0
env_settings:
num_envs: 1