Hugging Face's logo

hivex-research
/
hivex-DBR-PPO-baseline-task-6-difficulty-3

Reinforcement Learning
TensorBoard
ONNX
hivex
hivex-drone-based-reforestation
multi-agent-reinforcement-learning
Eval Results
Model card Files Metrics Community
philippds commited on
Commit
3f7ae1a
·
verified ·
1 Parent(s): e43fb3a

Upload 13 files

Browse files
Files changed (13) hide show
  1. Agent.onnx +3 -0
  2. Agent/Agent-1201032.onnx +3 -0
  3. Agent/Agent-1201032.pt +3 -0
  4. Agent/Agent-499998.onnx +3 -0
  5. Agent/Agent-499998.pt +3 -0
  6. Agent/Agent-999649.onnx +3 -0
  7. Agent/Agent-999649.pt +3 -0
  8. Agent/checkpoint.pt +3 -0
  9. Agent/events.out.tfevents.1717235416.RICHARD.26264.0 +3 -0
  10. README.md +39 -0
  11. configuration.yaml +107 -0
  12. run_logs/timers.json +587 -0
  13. run_logs/training_status.json +53 -0
Agent.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62da9fed4369edafc3357d617746f5607dec39315b0edad23faf69dd5d4fe6c8
3
+ size 562656
Agent/Agent-1201032.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62da9fed4369edafc3357d617746f5607dec39315b0edad23faf69dd5d4fe6c8
3
+ size 562656
Agent/Agent-1201032.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c29d8159be9b07a3ac02d4a33431cdf0218fffed4406194472f27346d89c11f
3
+ size 4530031
Agent/Agent-499998.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be92e5e1518c6515145b7454d8fd4aed4b4f5ddd22cd1bf06ed52163626263a3
3
+ size 562656
Agent/Agent-499998.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4262cc3999a03add0f432acf4da7dda1fd71f50df19c2c723cb014b1b01fbe98
3
+ size 4530031
Agent/Agent-999649.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69a977e43080c262e4905c1342b37e8ecf01277645270d7a2c4fc37e40d9ae30
3
+ size 562656
Agent/Agent-999649.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:069a5bf465f28193ee642139b3c638d0019d7c4c395e2e9b7b31762ccb2ee1a2
3
+ size 4530031
Agent/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c29d8159be9b07a3ac02d4a33431cdf0218fffed4406194472f27346d89c11f
3
+ size 4530031
Agent/events.out.tfevents.1717235416.RICHARD.26264.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1640c031c84203cdd975bd975681515885e9a1fce0a1775ccae6c724405fcba3
3
+ size 1856643
README.md ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: hivex
3
+ original_train_name: DroneBasedReforestation_difficulty_3_task_6_run_id_1_train
4
+ tags:
5
+ - hivex
6
+ - hivex-drone-based-reforestation
7
+ - reinforcement-learning
8
+ - multi-agent-reinforcement-learning
9
+ model-index:
10
+ - name: hivex-DBR-PPO-baseline-task-6-difficulty-3
11
+ results:
12
+ - task:
13
+ type: sub-task
14
+ name: explore_furthest_distance_and_return_to_base
15
+ task-id: 6
16
+ difficulty-id: 3
17
+ dataset:
18
+ name: hivex-drone-based-reforestation
19
+ type: hivex-drone-based-reforestation
20
+ metrics:
21
+ - type: furthest_distance_explored
22
+ value: 146.28668838500977 +/- 16.658292228774815
23
+ name: Furthest Distance Explored
24
+ verified: true
25
+ - type: out_of_energy_count
26
+ value: 0.595357158780098 +/- 0.08324645242738359
27
+ name: Out of Energy Count
28
+ verified: true
29
+ - type: recharge_energy_count
30
+ value: 131.29083390399813 +/- 117.44315350412963
31
+ name: Recharge Energy Count
32
+ verified: true
33
+ - type: cumulative_reward
34
+ value: 6.076253048032522 +/- 5.155621265658263
35
+ name: Cumulative Reward
36
+ verified: true
37
+ ---
38
+
39
+ This model serves as the baseline for the **Drone-Based Reforestation** environment, trained and tested on task <code>6</code> with difficulty <code>3</code> using the Proximal Policy Optimization (PPO) algorithm.<br><br>Environment: **Drone-Based Reforestation**<br>Task: <code>6</code><br>Difficulty: <code>3</code><br>Algorithm: <code>PPO</code><br>Episode Length: <code>2000</code><br>Training <code>max_steps</code>: <code>1200000</code><br>Testing <code>max_steps</code>: <code>300000</code><br><br>Train & Test [Scripts](https://github.com/hivex-research/hivex)<br>Download the [Environment](https://github.com/hivex-research/hivex-environments)
configuration.yaml ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ default_settings: null
2
+ behaviors:
3
+ Agent:
4
+ trainer_type: ppo
5
+ hyperparameters:
6
+ batch_size: 1024
7
+ buffer_size: 8192
8
+ learning_rate: 0.0003
9
+ beta: 0.005
10
+ epsilon: 0.2
11
+ lambd: 0.95
12
+ num_epoch: 3
13
+ shared_critic: false
14
+ learning_rate_schedule: linear
15
+ beta_schedule: linear
16
+ epsilon_schedule: linear
17
+ network_settings:
18
+ normalize: false
19
+ hidden_units: 128
20
+ num_layers: 2
21
+ vis_encode_type: resnet
22
+ memory: null
23
+ goal_conditioning_type: hyper
24
+ deterministic: false
25
+ reward_signals:
26
+ curiosity:
27
+ gamma: 0.99
28
+ strength: 0.1
29
+ network_settings:
30
+ normalize: false
31
+ hidden_units: 128
32
+ num_layers: 2
33
+ vis_encode_type: resnet
34
+ memory: null
35
+ goal_conditioning_type: hyper
36
+ deterministic: false
37
+ learning_rate: 0.0003
38
+ encoding_size: 256
39
+ extrinsic:
40
+ gamma: 0.99
41
+ strength: 0.9
42
+ network_settings:
43
+ normalize: false
44
+ hidden_units: 128
45
+ num_layers: 2
46
+ vis_encode_type: resnet
47
+ memory: null
48
+ goal_conditioning_type: hyper
49
+ deterministic: false
50
+ init_path: null
51
+ keep_checkpoints: 5
52
+ checkpoint_interval: 500000
53
+ max_steps: 1200000
54
+ time_horizon: 8192
55
+ summary_freq: 6000
56
+ threaded: true
57
+ self_play: null
58
+ behavioral_cloning: null
59
+ env_settings:
60
+ env_path: c:/users/pdsie/documents/hivex/src/hivex/training/baseline/ml_agents/dev_environments/Hivex_DroneBasedReforestation_win
61
+ env_args: null
62
+ base_port: 5007
63
+ num_envs: 1
64
+ num_areas: 1
65
+ seed: 5000
66
+ max_lifetime_restarts: 10
67
+ restarts_rate_limit_n: 1
68
+ restarts_rate_limit_period_s: 60
69
+ engine_settings:
70
+ width: 84
71
+ height: 84
72
+ quality_level: 5
73
+ time_scale: 20
74
+ target_frame_rate: -1
75
+ capture_frame_rate: 60
76
+ no_graphics: true
77
+ environment_parameters:
78
+ difficulty:
79
+ curriculum:
80
+ - value:
81
+ sampler_type: constant
82
+ sampler_parameters:
83
+ seed: 5000
84
+ value: 3
85
+ name: difficulty
86
+ completion_criteria: null
87
+ task:
88
+ curriculum:
89
+ - value:
90
+ sampler_type: constant
91
+ sampler_parameters:
92
+ seed: 5001
93
+ value: 6
94
+ name: task
95
+ completion_criteria: null
96
+ checkpoint_settings:
97
+ run_id: DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train
98
+ initialize_from: null
99
+ load_model: false
100
+ resume: false
101
+ force: false
102
+ train_model: false
103
+ inference: false
104
+ results_dir: results
105
+ torch_settings:
106
+ device: null
107
+ debug: false
run_logs/timers.json ADDED
@@ -0,0 +1,587 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "root",
3
+ "gauges": {
4
+ "Agent.Policy.Entropy.mean": {
5
+ "value": 1.4504915475845337,
6
+ "min": 1.4189385175704956,
7
+ "max": 1.4583607912063599,
8
+ "count": 200
9
+ },
10
+ "Agent.Policy.Entropy.sum": {
11
+ "value": 8716.00390625,
12
+ "min": 7019.41796875,
13
+ "max": 10183.787109375,
14
+ "count": 200
15
+ },
16
+ "Agent.DroneBasedReforestation.TreeDropCount.mean": {
17
+ "value": 0.0,
18
+ "min": 0.0,
19
+ "max": 0.0,
20
+ "count": 200
21
+ },
22
+ "Agent.DroneBasedReforestation.TreeDropCount.sum": {
23
+ "value": 0.0,
24
+ "min": 0.0,
25
+ "max": 0.0,
26
+ "count": 200
27
+ },
28
+ "Agent.DroneBasedReforestation.RechargeEnergyCount.mean": {
29
+ "value": 12.523809523809524,
30
+ "min": 0.0,
31
+ "max": 498.73333333333335,
32
+ "count": 200
33
+ },
34
+ "Agent.DroneBasedReforestation.RechargeEnergyCount.sum": {
35
+ "value": 263.0,
36
+ "min": 0.0,
37
+ "max": 7481.0,
38
+ "count": 200
39
+ },
40
+ "Agent.DroneBasedReforestation.SaveLocationCount.mean": {
41
+ "value": 0.0,
42
+ "min": 0.0,
43
+ "max": 0.0,
44
+ "count": 200
45
+ },
46
+ "Agent.DroneBasedReforestation.SaveLocationCount.sum": {
47
+ "value": 0.0,
48
+ "min": 0.0,
49
+ "max": 0.0,
50
+ "count": 200
51
+ },
52
+ "Agent.DroneBasedReforestation.OutofEnergyCount.mean": {
53
+ "value": 0.6666666666666666,
54
+ "min": 0.4,
55
+ "max": 0.8,
56
+ "count": 200
57
+ },
58
+ "Agent.DroneBasedReforestation.OutofEnergyCount.sum": {
59
+ "value": 14.0,
60
+ "min": 6.0,
61
+ "max": 14.0,
62
+ "count": 200
63
+ },
64
+ "Agent.DroneBasedReforestation.CumulativeDistanceUntilTreeDrop.mean": {
65
+ "value": 0.0,
66
+ "min": 0.0,
67
+ "max": 0.0,
68
+ "count": 200
69
+ },
70
+ "Agent.DroneBasedReforestation.CumulativeDistanceUntilTreeDrop.sum": {
71
+ "value": 0.0,
72
+ "min": 0.0,
73
+ "max": 0.0,
74
+ "count": 200
75
+ },
76
+ "Agent.DroneBasedReforestation.CumulativeTreeDropReward.mean": {
77
+ "value": 0.0,
78
+ "min": 0.0,
79
+ "max": 0.0,
80
+ "count": 200
81
+ },
82
+ "Agent.DroneBasedReforestation.CumulativeTreeDropReward.sum": {
83
+ "value": 0.0,
84
+ "min": 0.0,
85
+ "max": 0.0,
86
+ "count": 200
87
+ },
88
+ "Agent.DroneBasedReforestation.CumulativeDistanceReward.mean": {
89
+ "value": 0.0,
90
+ "min": 0.0,
91
+ "max": 0.0,
92
+ "count": 200
93
+ },
94
+ "Agent.DroneBasedReforestation.CumulativeDistanceReward.sum": {
95
+ "value": 0.0,
96
+ "min": 0.0,
97
+ "max": 0.0,
98
+ "count": 200
99
+ },
100
+ "Agent.DroneBasedReforestation.CumulativeNormalizedDistanceUntilTreeDrop.mean": {
101
+ "value": 0.0,
102
+ "min": 0.0,
103
+ "max": 0.0,
104
+ "count": 200
105
+ },
106
+ "Agent.DroneBasedReforestation.CumulativeNormalizedDistanceUntilTreeDrop.sum": {
107
+ "value": 0.0,
108
+ "min": 0.0,
109
+ "max": 0.0,
110
+ "count": 200
111
+ },
112
+ "Agent.DroneBasedReforestation.CumulativeDistancetoExistingTrees.mean": {
113
+ "value": 0.0,
114
+ "min": 0.0,
115
+ "max": 0.0,
116
+ "count": 200
117
+ },
118
+ "Agent.DroneBasedReforestation.CumulativeDistancetoExistingTrees.sum": {
119
+ "value": 0.0,
120
+ "min": 0.0,
121
+ "max": 0.0,
122
+ "count": 200
123
+ },
124
+ "Agent.DroneBasedReforestation.FurthestDistanceExplored.mean": {
125
+ "value": 130.73263047990343,
126
+ "min": 84.74893919626872,
127
+ "max": 187.240434773763,
128
+ "count": 200
129
+ },
130
+ "Agent.DroneBasedReforestation.FurthestDistanceExplored.sum": {
131
+ "value": 2745.3852400779724,
132
+ "min": 1525.480905532837,
133
+ "max": 3128.7300386428833,
134
+ "count": 200
135
+ },
136
+ "Agent.Environment.LessonNumber.difficulty.mean": {
137
+ "value": 0.0,
138
+ "min": 0.0,
139
+ "max": 0.0,
140
+ "count": 200
141
+ },
142
+ "Agent.Environment.LessonNumber.difficulty.sum": {
143
+ "value": 0.0,
144
+ "min": 0.0,
145
+ "max": 0.0,
146
+ "count": 200
147
+ },
148
+ "Agent.Environment.LessonNumber.task.mean": {
149
+ "value": 0.0,
150
+ "min": 0.0,
151
+ "max": 0.0,
152
+ "count": 200
153
+ },
154
+ "Agent.Environment.LessonNumber.task.sum": {
155
+ "value": 0.0,
156
+ "min": 0.0,
157
+ "max": 0.0,
158
+ "count": 200
159
+ },
160
+ "Agent.Environment.EpisodeLength.mean": {
161
+ "value": 332.6666666666667,
162
+ "min": 275.3333333333333,
163
+ "max": 399.0,
164
+ "count": 200
165
+ },
166
+ "Agent.Environment.EpisodeLength.sum": {
167
+ "value": 5988.0,
168
+ "min": 4788.0,
169
+ "max": 6990.0,
170
+ "count": 200
171
+ },
172
+ "Agent.Step.mean": {
173
+ "value": 1199832.0,
174
+ "min": 5600.0,
175
+ "max": 1199832.0,
176
+ "count": 200
177
+ },
178
+ "Agent.Step.sum": {
179
+ "value": 1199832.0,
180
+ "min": 5600.0,
181
+ "max": 1199832.0,
182
+ "count": 200
183
+ },
184
+ "Agent.Policy.CuriosityValueEstimate.mean": {
185
+ "value": 0.17566753923892975,
186
+ "min": 0.031902581453323364,
187
+ "max": 1.0969910621643066,
188
+ "count": 200
189
+ },
190
+ "Agent.Policy.CuriosityValueEstimate.sum": {
191
+ "value": 3.162015676498413,
192
+ "min": 0.4466361403465271,
193
+ "max": 18.22492027282715,
194
+ "count": 200
195
+ },
196
+ "Agent.Policy.ExtrinsicValueEstimate.mean": {
197
+ "value": 1.7508596181869507,
198
+ "min": -0.19481128454208374,
199
+ "max": 2.1962625980377197,
200
+ "count": 200
201
+ },
202
+ "Agent.Policy.ExtrinsicValueEstimate.sum": {
203
+ "value": 31.515472412109375,
204
+ "min": -3.5066030025482178,
205
+ "max": 33.99528121948242,
206
+ "count": 200
207
+ },
208
+ "Agent.Environment.CumulativeReward.mean": {
209
+ "value": 4.581625471667697,
210
+ "min": -1.0654666741689047,
211
+ "max": 20.056720465421677,
212
+ "count": 200
213
+ },
214
+ "Agent.Environment.CumulativeReward.sum": {
215
+ "value": 82.46925849001855,
216
+ "min": -15.98200011253357,
217
+ "max": 320.9075274467468,
218
+ "count": 200
219
+ },
220
+ "Agent.Policy.CuriosityReward.mean": {
221
+ "value": 0.5303110282661186,
222
+ "min": 0.0,
223
+ "max": 18.220220788319907,
224
+ "count": 200
225
+ },
226
+ "Agent.Policy.CuriosityReward.sum": {
227
+ "value": 9.545598508790135,
228
+ "min": 0.0,
229
+ "max": 273.3033118247986,
230
+ "count": 200
231
+ },
232
+ "Agent.Policy.ExtrinsicReward.mean": {
233
+ "value": 4.123461473474486,
234
+ "min": -0.9589201927185058,
235
+ "max": 18.051046889275312,
236
+ "count": 200
237
+ },
238
+ "Agent.Policy.ExtrinsicReward.sum": {
239
+ "value": 74.22230652254075,
240
+ "min": -14.383802890777588,
241
+ "max": 288.816750228405,
242
+ "count": 200
243
+ },
244
+ "Agent.IsTraining.mean": {
245
+ "value": 1.0,
246
+ "min": 1.0,
247
+ "max": 1.0,
248
+ "count": 200
249
+ },
250
+ "Agent.IsTraining.sum": {
251
+ "value": 1.0,
252
+ "min": 1.0,
253
+ "max": 1.0,
254
+ "count": 200
255
+ },
256
+ "Agent.Losses.PolicyLoss.mean": {
257
+ "value": 0.025146381948919345,
258
+ "min": 0.015833884438810248,
259
+ "max": 0.03159644683667769,
260
+ "count": 140
261
+ },
262
+ "Agent.Losses.PolicyLoss.sum": {
263
+ "value": 0.025146381948919345,
264
+ "min": 0.015833884438810248,
265
+ "max": 0.03159644683667769,
266
+ "count": 140
267
+ },
268
+ "Agent.Losses.ValueLoss.mean": {
269
+ "value": 7.141066680351893,
270
+ "min": 0.0002961420832434669,
271
+ "max": 18.593239585558575,
272
+ "count": 140
273
+ },
274
+ "Agent.Losses.ValueLoss.sum": {
275
+ "value": 7.141066680351893,
276
+ "min": 0.0002961420832434669,
277
+ "max": 18.593239585558575,
278
+ "count": 140
279
+ },
280
+ "Agent.Policy.LearningRate.mean": {
281
+ "value": 4.20999860000087e-08,
282
+ "min": 4.20999860000087e-08,
283
+ "max": 0.0002979000007,
284
+ "count": 140
285
+ },
286
+ "Agent.Policy.LearningRate.sum": {
287
+ "value": 4.20999860000087e-08,
288
+ "min": 4.20999860000087e-08,
289
+ "max": 0.0002979000007,
290
+ "count": 140
291
+ },
292
+ "Agent.Policy.Epsilon.mean": {
293
+ "value": 0.100014,
294
+ "min": 0.100014,
295
+ "max": 0.1993,
296
+ "count": 140
297
+ },
298
+ "Agent.Policy.Epsilon.sum": {
299
+ "value": 0.100014,
300
+ "min": 0.100014,
301
+ "max": 0.1993,
302
+ "count": 140
303
+ },
304
+ "Agent.Policy.Beta.mean": {
305
+ "value": 1.0698600000000144e-05,
306
+ "min": 1.0698600000000144e-05,
307
+ "max": 0.00496507,
308
+ "count": 140
309
+ },
310
+ "Agent.Policy.Beta.sum": {
311
+ "value": 1.0698600000000144e-05,
312
+ "min": 1.0698600000000144e-05,
313
+ "max": 0.00496507,
314
+ "count": 140
315
+ },
316
+ "Agent.Losses.CuriosityForwardLoss.mean": {
317
+ "value": 0.015817191878644127,
318
+ "min": 0.014498400889957944,
319
+ "max": 0.5835290277997652,
320
+ "count": 140
321
+ },
322
+ "Agent.Losses.CuriosityForwardLoss.sum": {
323
+ "value": 0.015817191878644127,
324
+ "min": 0.014498400889957944,
325
+ "max": 0.5835290277997652,
326
+ "count": 140
327
+ },
328
+ "Agent.Losses.CuriosityInverseLoss.mean": {
329
+ "value": 2.0491804033517838,
330
+ "min": 2.0173485726118088,
331
+ "max": 3.310828596353531,
332
+ "count": 140
333
+ },
334
+ "Agent.Losses.CuriosityInverseLoss.sum": {
335
+ "value": 2.0491804033517838,
336
+ "min": 2.0173485726118088,
337
+ "max": 3.310828596353531,
338
+ "count": 140
339
+ }
340
+ },
341
+ "metadata": {
342
+ "timer_format_version": "0.1.0",
343
+ "start_time_seconds": "1717235414",
344
+ "python_version": "3.9.18 (main, Sep 11 2023, 14:09:26) [MSC v.1916 64 bit (AMD64)]",
345
+ "command_line_arguments": "C:\\Users\\pdsie\\anaconda3\\envs\\mlagents20\\Scripts\\mlagents-learn c:/users/pdsie/documents/hivex/src/hivex/training/baseline/ml_agents/configs/mlagents/tmp/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train.yaml --run-id=DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train --base-port 5007",
346
+ "mlagents_version": "0.30.0",
347
+ "mlagents_envs_version": "0.30.0",
348
+ "communication_protocol_version": "1.5.0",
349
+ "pytorch_version": "1.7.1+cu110",
350
+ "numpy_version": "1.21.0",
351
+ "end_time_seconds": "1717239439"
352
+ },
353
+ "total": 4024.6476045,
354
+ "count": 1,
355
+ "self": 0.4105314000003091,
356
+ "children": {
357
+ "run_training.setup": {
358
+ "total": 0.05245299999999997,
359
+ "count": 1,
360
+ "self": 0.05245299999999997
361
+ },
362
+ "TrainerController.start_learning": {
363
+ "total": 4024.1846201,
364
+ "count": 1,
365
+ "self": 6.823291999885896,
366
+ "children": {
367
+ "TrainerController._reset_env": {
368
+ "total": 2.0034330000000002,
369
+ "count": 1,
370
+ "self": 2.0034330000000002
371
+ },
372
+ "TrainerController.advance": {
373
+ "total": 4015.132420500114,
374
+ "count": 401212,
375
+ "self": 6.283942800091609,
376
+ "children": {
377
+ "env_step": {
378
+ "total": 4008.8484777000226,
379
+ "count": 401212,
380
+ "self": 1688.6688614000836,
381
+ "children": {
382
+ "SubprocessEnvManager._take_step": {
383
+ "total": 2315.8741404999705,
384
+ "count": 401212,
385
+ "self": 12.286400099961611,
386
+ "children": {
387
+ "TorchPolicy.evaluate": {
388
+ "total": 2303.587740400009,
389
+ "count": 400346,
390
+ "self": 2303.587740400009
391
+ }
392
+ }
393
+ },
394
+ "workers": {
395
+ "total": 4.305475799968528,
396
+ "count": 401212,
397
+ "self": 0.0,
398
+ "children": {
399
+ "worker_root": {
400
+ "total": 4014.887415299974,
401
+ "count": 401212,
402
+ "is_parallel": true,
403
+ "self": 2570.5212920998615,
404
+ "children": {
405
+ "steps_from_proto": {
406
+ "total": 0.0062581999999999915,
407
+ "count": 1,
408
+ "is_parallel": true,
409
+ "self": 0.0001041000000001624,
410
+ "children": {
411
+ "_process_maybe_compressed_observation": {
412
+ "total": 0.006107399999999874,
413
+ "count": 2,
414
+ "is_parallel": true,
415
+ "self": 2.9700000000021376e-05,
416
+ "children": {
417
+ "_observation_to_np_array": {
418
+ "total": 0.006077699999999853,
419
+ "count": 3,
420
+ "is_parallel": true,
421
+ "self": 2.819999999981171e-05,
422
+ "children": {
423
+ "process_pixels": {
424
+ "total": 0.006049500000000041,
425
+ "count": 3,
426
+ "is_parallel": true,
427
+ "self": 0.0002406999999999826,
428
+ "children": {
429
+ "image_decompress": {
430
+ "total": 0.005808800000000058,
431
+ "count": 3,
432
+ "is_parallel": true,
433
+ "self": 0.005808800000000058
434
+ }
435
+ }
436
+ }
437
+ }
438
+ }
439
+ }
440
+ },
441
+ "_process_rank_one_or_two_observation": {
442
+ "total": 4.669999999995511e-05,
443
+ "count": 2,
444
+ "is_parallel": true,
445
+ "self": 4.669999999995511e-05
446
+ }
447
+ }
448
+ },
449
+ "UnityEnvironment.step": {
450
+ "total": 1444.3598650001122,
451
+ "count": 401212,
452
+ "is_parallel": true,
453
+ "self": 22.982416200152784,
454
+ "children": {
455
+ "UnityEnvironment._generate_step_input": {
456
+ "total": 24.666361000051545,
457
+ "count": 401212,
458
+ "is_parallel": true,
459
+ "self": 24.666361000051545
460
+ },
461
+ "communicator.exchange": {
462
+ "total": 1209.1215095999087,
463
+ "count": 401212,
464
+ "is_parallel": true,
465
+ "self": 1209.1215095999087
466
+ },
467
+ "steps_from_proto": {
468
+ "total": 187.58957819999944,
469
+ "count": 401212,
470
+ "is_parallel": true,
471
+ "self": 37.552167899814066,
472
+ "children": {
473
+ "_process_maybe_compressed_observation": {
474
+ "total": 134.0370428000622,
475
+ "count": 802424,
476
+ "is_parallel": true,
477
+ "self": 10.410967999887362,
478
+ "children": {
479
+ "_observation_to_np_array": {
480
+ "total": 123.62607480017485,
481
+ "count": 1204305,
482
+ "is_parallel": true,
483
+ "self": 10.863587100228912,
484
+ "children": {
485
+ "process_pixels": {
486
+ "total": 112.76248769994594,
487
+ "count": 1204305,
488
+ "is_parallel": true,
489
+ "self": 53.50085809998973,
490
+ "children": {
491
+ "image_decompress": {
492
+ "total": 59.26162959995621,
493
+ "count": 1204305,
494
+ "is_parallel": true,
495
+ "self": 59.26162959995621
496
+ }
497
+ }
498
+ }
499
+ }
500
+ }
501
+ }
502
+ },
503
+ "_process_rank_one_or_two_observation": {
504
+ "total": 16.000367500123158,
505
+ "count": 802424,
506
+ "is_parallel": true,
507
+ "self": 16.000367500123158
508
+ }
509
+ }
510
+ }
511
+ }
512
+ }
513
+ }
514
+ }
515
+ }
516
+ }
517
+ }
518
+ }
519
+ }
520
+ },
521
+ "trainer_threads": {
522
+ "total": 3.2500000088475645e-05,
523
+ "count": 1,
524
+ "self": 3.2500000088475645e-05,
525
+ "children": {
526
+ "thread_root": {
527
+ "total": 0.0,
528
+ "count": 0,
529
+ "is_parallel": true,
530
+ "self": 0.0,
531
+ "children": {
532
+ "trainer_advance": {
533
+ "total": 4020.277517000015,
534
+ "count": 191258,
535
+ "is_parallel": true,
536
+ "self": 5.217701400116312,
537
+ "children": {
538
+ "process_trajectory": {
539
+ "total": 3147.2481361998975,
540
+ "count": 191258,
541
+ "is_parallel": true,
542
+ "self": 3146.6857697998976,
543
+ "children": {
544
+ "RLTrainer._checkpoint": {
545
+ "total": 0.5623664000002009,
546
+ "count": 2,
547
+ "is_parallel": true,
548
+ "self": 0.5623664000002009
549
+ }
550
+ }
551
+ },
552
+ "_update_policy": {
553
+ "total": 867.8116794000011,
554
+ "count": 140,
555
+ "is_parallel": true,
556
+ "self": 579.4782453999968,
557
+ "children": {
558
+ "TorchPPOOptimizer.update": {
559
+ "total": 288.33343400000433,
560
+ "count": 3384,
561
+ "is_parallel": true,
562
+ "self": 288.33343400000433
563
+ }
564
+ }
565
+ }
566
+ }
567
+ }
568
+ }
569
+ }
570
+ }
571
+ },
572
+ "TrainerController._save_models": {
573
+ "total": 0.22544209999978193,
574
+ "count": 1,
575
+ "self": 0.007709299999532959,
576
+ "children": {
577
+ "RLTrainer._checkpoint": {
578
+ "total": 0.21773280000024897,
579
+ "count": 1,
580
+ "self": 0.21773280000024897
581
+ }
582
+ }
583
+ }
584
+ }
585
+ }
586
+ }
587
+ }
run_logs/training_status.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "difficulty": {
3
+ "lesson_num": 0
4
+ },
5
+ "task": {
6
+ "lesson_num": 0
7
+ },
8
+ "Agent": {
9
+ "checkpoints": [
10
+ {
11
+ "steps": 499998,
12
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-499998.onnx",
13
+ "reward": 4.425038735071818,
14
+ "creation_time": 1717237090.5635085,
15
+ "auxillary_file_paths": [
16
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-499998.pt"
17
+ ]
18
+ },
19
+ {
20
+ "steps": 999649,
21
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-999649.onnx",
22
+ "reward": 0.5965502442497956,
23
+ "creation_time": 1717238763.9110703,
24
+ "auxillary_file_paths": [
25
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-999649.pt"
26
+ ]
27
+ },
28
+ {
29
+ "steps": 1201032,
30
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-1201032.onnx",
31
+ "reward": -0.9994999766349792,
32
+ "creation_time": 1717239439.1295176,
33
+ "auxillary_file_paths": [
34
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-1201032.pt"
35
+ ]
36
+ }
37
+ ],
38
+ "final_checkpoint": {
39
+ "steps": 1201032,
40
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent.onnx",
41
+ "reward": -0.9994999766349792,
42
+ "creation_time": 1717239439.1295176,
43
+ "auxillary_file_paths": [
44
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_3_task_6_run_id_1_train\\Agent\\Agent-1201032.pt"
45
+ ]
46
+ }
47
+ },
48
+ "metadata": {
49
+ "stats_format_version": "0.3.0",
50
+ "mlagents_version": "0.30.0",
51
+ "torch_version": "1.7.1+cu110"
52
+ }
53
+ }