/
sac_trainer_config.yaml
268 lines (247 loc) · 5.11 KB
/
sac_trainer_config.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
default:
trainer: sac
batch_size: 128
buffer_size: 50000
buffer_init_steps: 0
hidden_units: 128
init_entcoef: 1.0
learning_rate: 3.0e-4
learning_rate_schedule: constant
max_steps: 5.0e4
memory_size: 256
normalize: false
num_update: 1
train_interval: 1
num_layers: 2
time_horizon: 64
sequence_length: 64
summary_freq: 1000
tau: 0.005
use_recurrent: false
vis_encode_type: simple
reward_signals:
extrinsic:
strength: 1.0
gamma: 0.99
FoodCollector:
normalize: false
batch_size: 256
buffer_size: 500000
max_steps: 1.0e5
init_entcoef: 0.05
train_interval: 1
Bouncer:
normalize: true
max_steps: 5.0e5
num_layers: 2
hidden_units: 64
summary_freq: 1000
PushBlock:
max_steps: 5.0e4
init_entcoef: 0.05
hidden_units: 256
summary_freq: 2000
time_horizon: 64
num_layers: 2
SmallWallJump:
max_steps: 1.0e6
hidden_units: 256
summary_freq: 2000
time_horizon: 128
init_entcoef: 0.1
num_layers: 2
normalize: false
BigWallJump:
max_steps: 1.0e6
hidden_units: 256
summary_freq: 2000
time_horizon: 128
num_layers: 2
init_entcoef: 0.1
normalize: false
Striker:
max_steps: 5.0e5
learning_rate: 1e-3
hidden_units: 256
summary_freq: 2000
time_horizon: 128
init_entcoef: 0.1
num_layers: 2
normalize: false
Goalie:
max_steps: 5.0e5
learning_rate: 1e-3
hidden_units: 256
summary_freq: 2000
time_horizon: 128
init_entcoef: 0.1
num_layers: 2
normalize: false
Pyramids:
summary_freq: 2000
time_horizon: 128
batch_size: 128
buffer_init_steps: 10000
buffer_size: 500000
hidden_units: 256
num_layers: 2
init_entcoef: 0.01
max_steps: 5.0e5
sequence_length: 16
tau: 0.01
use_recurrent: false
reward_signals:
extrinsic:
strength: 2.0
gamma: 0.99
gail:
strength: 0.02
gamma: 0.99
encoding_size: 128
use_actions: true
demo_path: demos/ExpertPyramid.demo
VisualPyramids:
time_horizon: 128
batch_size: 64
hidden_units: 256
buffer_init_steps: 1000
num_layers: 1
max_steps: 5.0e5
buffer_size: 500000
init_entcoef: 0.01
tau: 0.01
reward_signals:
extrinsic:
strength: 2.0
gamma: 0.99
gail:
strength: 0.02
gamma: 0.99
encoding_size: 128
use_actions: true
demo_path: demos/ExpertPyramid.demo
3DBall:
normalize: true
batch_size: 64
buffer_size: 12000
summary_freq: 1000
time_horizon: 1000
hidden_units: 64
init_entcoef: 0.5
3DBallHard:
normalize: true
batch_size: 256
summary_freq: 1000
time_horizon: 1000
Tennis:
buffer_size: 500000
normalize: true
max_steps: 2e5
CrawlerStatic:
normalize: true
time_horizon: 1000
batch_size: 256
train_interval: 2
buffer_size: 500000
buffer_init_steps: 2000
max_steps: 5e5
summary_freq: 3000
init_entcoef: 1.0
num_layers: 3
hidden_units: 512
reward_signals:
extrinsic:
strength: 1.0
gamma: 0.995
CrawlerDynamic:
normalize: true
time_horizon: 1000
batch_size: 256
buffer_size: 500000
summary_freq: 3000
train_interval: 2
num_layers: 3
max_steps: 1e6
hidden_units: 512
reward_signals:
extrinsic:
strength: 1.0
gamma: 0.995
Walker:
normalize: true
time_horizon: 1000
batch_size: 256
buffer_size: 500000
max_steps: 2e6
summary_freq: 3000
num_layers: 4
train_interval: 2
hidden_units: 512
reward_signals:
extrinsic:
strength: 1.0
gamma: 0.995
Reacher:
normalize: true
time_horizon: 1000
batch_size: 128
buffer_size: 500000
max_steps: 2e5
summary_freq: 3000
Hallway:
sequence_length: 32
num_layers: 2
hidden_units: 128
memory_size: 256
init_entcoef: 0.1
max_steps: 5.0e5
summary_freq: 1000
time_horizon: 64
use_recurrent: true
VisualHallway:
sequence_length: 32
num_layers: 1
hidden_units: 128
memory_size: 256
gamma: 0.99
batch_size: 64
max_steps: 5.0e5
summary_freq: 1000
time_horizon: 64
use_recurrent: true
VisualPushBlock:
use_recurrent: true
sequence_length: 32
num_layers: 1
hidden_units: 128
memory_size: 256
gamma: 0.99
buffer_size: 1024
batch_size: 64
max_steps: 5.0e5
summary_freq: 1000
time_horizon: 64
GridWorld:
batch_size: 128
normalize: false
num_layers: 1
hidden_units: 128
init_entcoef: 0.5
buffer_init_steps: 1000
buffer_size: 50000
max_steps: 50000
summary_freq: 2000
time_horizon: 5
reward_signals:
extrinsic:
strength: 1.0
gamma: 0.9
Basic:
batch_size: 64
normalize: false
num_layers: 2
init_entcoef: 0.01
hidden_units: 20
max_steps: 5.0e5
summary_freq: 2000
time_horizon: 10