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
25 行
536 B
25 行
536 B
behaviors:
|
|
Reacher:
|
|
trainer: sac
|
|
batch_size: 128
|
|
buffer_size: 500000
|
|
buffer_init_steps: 0
|
|
hidden_units: 128
|
|
init_entcoef: 1.0
|
|
learning_rate: 0.0003
|
|
learning_rate_schedule: constant
|
|
max_steps: 2e7
|
|
memory_size: 128
|
|
normalize: true
|
|
steps_per_update: 20
|
|
num_layers: 2
|
|
time_horizon: 1000
|
|
sequence_length: 64
|
|
summary_freq: 60000
|
|
tau: 0.005
|
|
use_recurrent: false
|
|
vis_encode_type: simple
|
|
reward_signals:
|
|
extrinsic:
|
|
strength: 1.0
|
|
gamma: 0.99
|