Hugging Face's logo

hivex-research
/
hivex-DBR-PPO-baseline-task-3-difficulty-7

Reinforcement Learning
TensorBoard
ONNX
hivex
hivex-drone-based-reforestation
multi-agent-reinforcement-learning
Eval Results
Model card Files Metrics Community
philippds commited on
Commit
7428b32
·
verified ·
1 Parent(s): f065710

Upload 13 files

Browse files
Files changed (13) hide show
  1. Agent.onnx +3 -0
  2. Agent/Agent-1201044.onnx +3 -0
  3. Agent/Agent-1201044.pt +3 -0
  4. Agent/Agent-499998.onnx +3 -0
  5. Agent/Agent-499998.pt +3 -0
  6. Agent/Agent-999997.onnx +3 -0
  7. Agent/Agent-999997.pt +3 -0
  8. Agent/checkpoint.pt +3 -0
  9. Agent/events.out.tfevents.1717702125.RICHARD.39220.0 +3 -0
  10. README.md +59 -0
  11. configuration.yaml +107 -0
  12. run_logs/timers.json +575 -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:ca537ac9235d3fa1c79e321ac7934671218b72f0bd207e81995e0a1509e7ef3f
3
+ size 562656
Agent/Agent-1201044.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca537ac9235d3fa1c79e321ac7934671218b72f0bd207e81995e0a1509e7ef3f
3
+ size 562656
Agent/Agent-1201044.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30594d53fa8b2f54a00db3f6d448ce75a859b87e53dc24d5de2f23211e0e30d2
3
+ size 4530031
Agent/Agent-499998.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3903d02f8e0ecdfc4475f84357563d6ed724ee8f233cf14b581e37f9ef433e2e
3
+ size 562656
Agent/Agent-499998.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddac82ff7b314ee79bc9f1a8517f172e98e91a23f1ca764362f93816d94898ba
3
+ size 4530031
Agent/Agent-999997.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc98a1ed9afcbaa8bf0e9b5c2c8df3739a37b805914d8e9b7b0b38ecdd4caef9
3
+ size 562656
Agent/Agent-999997.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0b2208549483f4bda95a717479a69ac03b06386675b93cae4be916ae56ea2fd
3
+ size 4530031
Agent/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30594d53fa8b2f54a00db3f6d448ce75a859b87e53dc24d5de2f23211e0e30d2
3
+ size 4530031
Agent/events.out.tfevents.1717702125.RICHARD.39220.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d2907890f0e364a5452c37fe1d86af85b78e8fd7141d92163a7f092b0a18078
3
+ size 2389739
README.md ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: hivex
3
+ original_train_name: DroneBasedReforestation_difficulty_7_task_3_run_id_2_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-3-difficulty-7
11
+ results:
12
+ - task:
13
+ type: sub-task
14
+ name: drop_seed
15
+ task-id: 3
16
+ difficulty-id: 7
17
+ dataset:
18
+ name: hivex-drone-based-reforestation
19
+ type: hivex-drone-based-reforestation
20
+ metrics:
21
+ - type: cumulative_distance_reward
22
+ value: 1.3364013075828551 +/- 0.28799869386679394
23
+ name: Cumulative Distance Reward
24
+ verified: true
25
+ - type: cumulative_distance_until_tree_drop
26
+ value: 49.32145935058594 +/- 5.925578414273251
27
+ name: Cumulative Distance Until Tree Drop
28
+ verified: true
29
+ - type: cumulative_distance_to_existing_trees
30
+ value: 65.15668449401855 +/- 5.863480554164979
31
+ name: Cumulative Distance to Existing Trees
32
+ verified: true
33
+ - type: cumulative_normalized_distance_until_tree_drop
34
+ value: 0.1336401303112507 +/- 0.028799868452171418
35
+ name: Cumulative Normalized Distance Until Tree Drop
36
+ verified: true
37
+ - type: cumulative_tree_drop_reward
38
+ value: 3.9665349340438842 +/- 0.7359914045841582
39
+ name: Cumulative Tree Drop Reward
40
+ verified: true
41
+ - type: out_of_energy_count
42
+ value: 0.03556458073668182 +/- 0.01864171812248584
43
+ name: Out of Energy Count
44
+ verified: true
45
+ - type: recharge_energy_count
46
+ value: 11.073680667877197 +/- 0.6686583978261852
47
+ name: Recharge Energy Count
48
+ verified: true
49
+ - type: tree_drop_count
50
+ value: 0.9511495363712311 +/- 0.027165762086272822
51
+ name: Tree Drop Count
52
+ verified: true
53
+ - type: cumulative_reward
54
+ value: 101.46505523681641 +/- 3.079002322895952
55
+ name: Cumulative Reward
56
+ verified: true
57
+ ---
58
+
59
+ This model serves as the baseline for the **Drone-Based Reforestation** environment, trained and tested on task <code>3</code> with difficulty <code>7</code> using the Proximal Policy Optimization (PPO) algorithm.<br><br>Environment: **Drone-Based Reforestation**<br>Task: <code>3</code><br>Difficulty: <code>7</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: 7
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: 3
94
+ name: task
95
+ completion_criteria: null
96
+ checkpoint_settings:
97
+ run_id: DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_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 +1,575 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "root",
3
+ "gauges": {
4
+ "Agent.Policy.Entropy.mean": {
5
+ "value": 1.3848451375961304,
6
+ "min": 1.3848425149917603,
7
+ "max": 1.4281865358352661,
8
+ "count": 200
9
+ },
10
+ "Agent.Policy.Entropy.sum": {
11
+ "value": 9721.61328125,
12
+ "min": 6707.294921875,
13
+ "max": 9935.82421875,
14
+ "count": 200
15
+ },
16
+ "Agent.DroneBasedReforestation.TreeDropCount.mean": {
17
+ "value": 0.9130434782608695,
18
+ "min": 0.3333333333333333,
19
+ "max": 0.991869918699187,
20
+ "count": 200
21
+ },
22
+ "Agent.DroneBasedReforestation.TreeDropCount.sum": {
23
+ "value": 63.0,
24
+ "min": 5.0,
25
+ "max": 122.0,
26
+ "count": 200
27
+ },
28
+ "Agent.DroneBasedReforestation.RechargeEnergyCount.mean": {
29
+ "value": 11.057971014492754,
30
+ "min": 2.2,
31
+ "max": 44.8,
32
+ "count": 200
33
+ },
34
+ "Agent.DroneBasedReforestation.RechargeEnergyCount.sum": {
35
+ "value": 763.0,
36
+ "min": 33.0,
37
+ "max": 1486.0,
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": 0.07246376811594203,
54
+ "min": 0.008130081300813009,
55
+ "max": 0.5333333333333333,
56
+ "count": 200
57
+ },
58
+ "Agent.DroneBasedReforestation.OutofEnergyCount.sum": {
59
+ "value": 5.0,
60
+ "min": 1.0,
61
+ "max": 12.0,
62
+ "count": 200
63
+ },
64
+ "Agent.DroneBasedReforestation.CumulativeDistanceUntilTreeDrop.mean": {
65
+ "value": 52.58073242850926,
66
+ "min": 12.207405726114908,
67
+ "max": 61.72031497955322,
68
+ "count": 200
69
+ },
70
+ "Agent.DroneBasedReforestation.CumulativeDistanceUntilTreeDrop.sum": {
71
+ "value": 3628.0705375671387,
72
+ "min": 250.79424285888672,
73
+ "max": 5155.910724639893,
74
+ "count": 200
75
+ },
76
+ "Agent.DroneBasedReforestation.CumulativeTreeDropReward.mean": {
77
+ "value": 3.26475090150168,
78
+ "min": 0.07396268049875895,
79
+ "max": 6.10832218436355,
80
+ "count": 200
81
+ },
82
+ "Agent.DroneBasedReforestation.CumulativeTreeDropReward.sum": {
83
+ "value": 225.2678122036159,
84
+ "min": 1.1094402074813843,
85
+ "max": 572.8081967830658,
86
+ "count": 200
87
+ },
88
+ "Agent.DroneBasedReforestation.CumulativeDistanceReward.mean": {
89
+ "value": 1.3341534353684688,
90
+ "min": 0.2047844628492991,
91
+ "max": 2.016074032584826,
92
+ "count": 200
93
+ },
94
+ "Agent.DroneBasedReforestation.CumulativeDistanceReward.sum": {
95
+ "value": 92.05658704042435,
96
+ "min": 3.5308263301849365,
97
+ "max": 159.98608255386353,
98
+ "count": 200
99
+ },
100
+ "Agent.DroneBasedReforestation.CumulativeNormalizedDistanceUntilTreeDrop.mean": {
101
+ "value": 0.13341534483260004,
102
+ "min": 0.020478446036577225,
103
+ "max": 0.20160740353167056,
104
+ "count": 200
105
+ },
106
+ "Agent.DroneBasedReforestation.CumulativeNormalizedDistanceUntilTreeDrop.sum": {
107
+ "value": 9.205658793449402,
108
+ "min": 0.3530826270580292,
109
+ "max": 15.998608238995075,
110
+ "count": 200
111
+ },
112
+ "Agent.DroneBasedReforestation.CumulativeDistancetoExistingTrees.mean": {
113
+ "value": 65.95886754125789,
114
+ "min": 13.084621217515734,
115
+ "max": 77.89113126595815,
116
+ "count": 200
117
+ },
118
+ "Agent.DroneBasedReforestation.CumulativeDistancetoExistingTrees.sum": {
119
+ "value": 4551.161860346794,
120
+ "min": 235.5231819152832,
121
+ "max": 8497.22591650486,
122
+ "count": 200
123
+ },
124
+ "Agent.Environment.LessonNumber.difficulty.mean": {
125
+ "value": 0.0,
126
+ "min": 0.0,
127
+ "max": 0.0,
128
+ "count": 200
129
+ },
130
+ "Agent.Environment.LessonNumber.difficulty.sum": {
131
+ "value": 0.0,
132
+ "min": 0.0,
133
+ "max": 0.0,
134
+ "count": 200
135
+ },
136
+ "Agent.Environment.LessonNumber.task.mean": {
137
+ "value": 0.0,
138
+ "min": 0.0,
139
+ "max": 0.0,
140
+ "count": 200
141
+ },
142
+ "Agent.Environment.LessonNumber.task.sum": {
143
+ "value": 0.0,
144
+ "min": 0.0,
145
+ "max": 0.0,
146
+ "count": 200
147
+ },
148
+ "Agent.Environment.EpisodeLength.mean": {
149
+ "value": 100.73913043478261,
150
+ "min": 45.36585365853659,
151
+ "max": 399.0,
152
+ "count": 200
153
+ },
154
+ "Agent.Environment.EpisodeLength.sum": {
155
+ "value": 6951.0,
156
+ "min": 4746.0,
157
+ "max": 7038.0,
158
+ "count": 200
159
+ },
160
+ "Agent.Step.mean": {
161
+ "value": 1199844.0,
162
+ "min": 5673.0,
163
+ "max": 1199844.0,
164
+ "count": 200
165
+ },
166
+ "Agent.Step.sum": {
167
+ "value": 1199844.0,
168
+ "min": 5673.0,
169
+ "max": 1199844.0,
170
+ "count": 200
171
+ },
172
+ "Agent.Policy.CuriosityValueEstimate.mean": {
173
+ "value": 0.24977338314056396,
174
+ "min": 0.013205585069954395,
175
+ "max": 1.064765453338623,
176
+ "count": 200
177
+ },
178
+ "Agent.Policy.CuriosityValueEstimate.sum": {
179
+ "value": 17.234363555908203,
180
+ "min": 0.3961675465106964,
181
+ "max": 34.85499572753906,
182
+ "count": 200
183
+ },
184
+ "Agent.Policy.ExtrinsicValueEstimate.mean": {
185
+ "value": 50.36190414428711,
186
+ "min": 0.05967150628566742,
187
+ "max": 53.74601364135742,
188
+ "count": 200
189
+ },
190
+ "Agent.Policy.ExtrinsicValueEstimate.sum": {
191
+ "value": 3474.971435546875,
192
+ "min": 1.4917876720428467,
193
+ "max": 5976.26953125,
194
+ "count": 200
195
+ },
196
+ "Agent.Environment.CumulativeReward.mean": {
197
+ "value": 100.01051464957604,
198
+ "min": 33.38887961308161,
199
+ "max": 106.3299504097062,
200
+ "count": 200
201
+ },
202
+ "Agent.Environment.CumulativeReward.sum": {
203
+ "value": 6900.725510820746,
204
+ "min": 500.8331941962242,
205
+ "max": 12919.049165844917,
206
+ "count": 200
207
+ },
208
+ "Agent.Policy.CuriosityReward.mean": {
209
+ "value": 0.20973035077685895,
210
+ "min": 0.0,
211
+ "max": 5.984334189551217,
212
+ "count": 200
213
+ },
214
+ "Agent.Policy.CuriosityReward.sum": {
215
+ "value": 14.471394203603268,
216
+ "min": 0.0,
217
+ "max": 209.4516966342926,
218
+ "count": 200
219
+ },
220
+ "Agent.Policy.ExtrinsicReward.mean": {
221
+ "value": 90.00946634943071,
222
+ "min": 30.04999473889669,
223
+ "max": 95.69695790969965,
224
+ "count": 200
225
+ },
226
+ "Agent.Policy.ExtrinsicReward.sum": {
227
+ "value": 6210.653178110719,
228
+ "min": 450.7499210834503,
229
+ "max": 11627.144557833672,
230
+ "count": 200
231
+ },
232
+ "Agent.IsTraining.mean": {
233
+ "value": 1.0,
234
+ "min": 1.0,
235
+ "max": 1.0,
236
+ "count": 200
237
+ },
238
+ "Agent.IsTraining.sum": {
239
+ "value": 1.0,
240
+ "min": 1.0,
241
+ "max": 1.0,
242
+ "count": 200
243
+ },
244
+ "Agent.Losses.PolicyLoss.mean": {
245
+ "value": 0.026116722729057074,
246
+ "min": 0.014573930141826471,
247
+ "max": 0.03793451966096958,
248
+ "count": 138
249
+ },
250
+ "Agent.Losses.PolicyLoss.sum": {
251
+ "value": 0.026116722729057074,
252
+ "min": 0.014573930141826471,
253
+ "max": 0.03793451966096958,
254
+ "count": 138
255
+ },
256
+ "Agent.Losses.ValueLoss.mean": {
257
+ "value": 97.28924910227458,
258
+ "min": 22.247718532880146,
259
+ "max": 144.694242477417,
260
+ "count": 138
261
+ },
262
+ "Agent.Losses.ValueLoss.sum": {
263
+ "value": 97.28924910227458,
264
+ "min": 22.247718532880146,
265
+ "max": 144.694242477417,
266
+ "count": 138
267
+ },
268
+ "Agent.Policy.LearningRate.mean": {
269
+ "value": 8.265997244999856e-07,
270
+ "min": 8.265997244999856e-07,
271
+ "max": 0.00029776350074549996,
272
+ "count": 138
273
+ },
274
+ "Agent.Policy.LearningRate.sum": {
275
+ "value": 8.265997244999856e-07,
276
+ "min": 8.265997244999856e-07,
277
+ "max": 0.00029776350074549996,
278
+ "count": 138
279
+ },
280
+ "Agent.Policy.Epsilon.mean": {
281
+ "value": 0.1002755,
282
+ "min": 0.1002755,
283
+ "max": 0.1992545,
284
+ "count": 138
285
+ },
286
+ "Agent.Policy.Epsilon.sum": {
287
+ "value": 0.1002755,
288
+ "min": 0.1002755,
289
+ "max": 0.1992545,
290
+ "count": 138
291
+ },
292
+ "Agent.Policy.Beta.mean": {
293
+ "value": 2.3747449999999762e-05,
294
+ "min": 2.3747449999999762e-05,
295
+ "max": 0.00496279955,
296
+ "count": 138
297
+ },
298
+ "Agent.Policy.Beta.sum": {
299
+ "value": 2.3747449999999762e-05,
300
+ "min": 2.3747449999999762e-05,
301
+ "max": 0.00496279955,
302
+ "count": 138
303
+ },
304
+ "Agent.Losses.CuriosityForwardLoss.mean": {
305
+ "value": 0.025772807653993368,
306
+ "min": 0.018984782819946606,
307
+ "max": 0.6002316027879715,
308
+ "count": 138
309
+ },
310
+ "Agent.Losses.CuriosityForwardLoss.sum": {
311
+ "value": 0.025772807653993368,
312
+ "min": 0.018984782819946606,
313
+ "max": 0.6002316027879715,
314
+ "count": 138
315
+ },
316
+ "Agent.Losses.CuriosityInverseLoss.mean": {
317
+ "value": 2.3072385787963867,
318
+ "min": 2.2800092498461404,
319
+ "max": 3.353024572134018,
320
+ "count": 138
321
+ },
322
+ "Agent.Losses.CuriosityInverseLoss.sum": {
323
+ "value": 2.3072385787963867,
324
+ "min": 2.2800092498461404,
325
+ "max": 3.353024572134018,
326
+ "count": 138
327
+ }
328
+ },
329
+ "metadata": {
330
+ "timer_format_version": "0.1.0",
331
+ "start_time_seconds": "1717702123",
332
+ "python_version": "3.9.18 (main, Sep 11 2023, 14:09:26) [MSC v.1916 64 bit (AMD64)]",
333
+ "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_7_task_3_run_id_2_train.yaml --run-id=DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train --base-port 5007",
334
+ "mlagents_version": "0.30.0",
335
+ "mlagents_envs_version": "0.30.0",
336
+ "communication_protocol_version": "1.5.0",
337
+ "pytorch_version": "1.7.1+cu110",
338
+ "numpy_version": "1.21.0",
339
+ "end_time_seconds": "1717706411"
340
+ },
341
+ "total": 4288.1489367,
342
+ "count": 1,
343
+ "self": 0.49392859999989014,
344
+ "children": {
345
+ "run_training.setup": {
346
+ "total": 0.04943300000000006,
347
+ "count": 1,
348
+ "self": 0.04943300000000006
349
+ },
350
+ "TrainerController.start_learning": {
351
+ "total": 4287.6055751,
352
+ "count": 1,
353
+ "self": 5.63005070006875,
354
+ "children": {
355
+ "TrainerController._reset_env": {
356
+ "total": 2.0493757,
357
+ "count": 1,
358
+ "self": 2.0493757
359
+ },
360
+ "TrainerController.advance": {
361
+ "total": 4279.68292289993,
362
+ "count": 403195,
363
+ "self": 5.4479560000054335,
364
+ "children": {
365
+ "env_step": {
366
+ "total": 4274.234966899925,
367
+ "count": 403195,
368
+ "self": 2056.9615485001286,
369
+ "children": {
370
+ "SubprocessEnvManager._take_step": {
371
+ "total": 2213.2936437999447,
372
+ "count": 403195,
373
+ "self": 11.579975600094258,
374
+ "children": {
375
+ "TorchPolicy.evaluate": {
376
+ "total": 2201.7136681998504,
377
+ "count": 400349,
378
+ "self": 2201.7136681998504
379
+ }
380
+ }
381
+ },
382
+ "workers": {
383
+ "total": 3.9797745998515874,
384
+ "count": 403195,
385
+ "self": 0.0,
386
+ "children": {
387
+ "worker_root": {
388
+ "total": 4278.062895000038,
389
+ "count": 403195,
390
+ "is_parallel": true,
391
+ "self": 2467.8336563999337,
392
+ "children": {
393
+ "steps_from_proto": {
394
+ "total": 0.006490000000000107,
395
+ "count": 1,
396
+ "is_parallel": true,
397
+ "self": 0.00010290000000034993,
398
+ "children": {
399
+ "_process_maybe_compressed_observation": {
400
+ "total": 0.006341599999999836,
401
+ "count": 2,
402
+ "is_parallel": true,
403
+ "self": 3.299999999950565e-05,
404
+ "children": {
405
+ "_observation_to_np_array": {
406
+ "total": 0.006308600000000331,
407
+ "count": 3,
408
+ "is_parallel": true,
409
+ "self": 2.990000000013815e-05,
410
+ "children": {
411
+ "process_pixels": {
412
+ "total": 0.0062787000000001925,
413
+ "count": 3,
414
+ "is_parallel": true,
415
+ "self": 0.00024170000000034442,
416
+ "children": {
417
+ "image_decompress": {
418
+ "total": 0.006036999999999848,
419
+ "count": 3,
420
+ "is_parallel": true,
421
+ "self": 0.006036999999999848
422
+ }
423
+ }
424
+ }
425
+ }
426
+ }
427
+ }
428
+ },
429
+ "_process_rank_one_or_two_observation": {
430
+ "total": 4.5499999999920604e-05,
431
+ "count": 2,
432
+ "is_parallel": true,
433
+ "self": 4.5499999999920604e-05
434
+ }
435
+ }
436
+ },
437
+ "UnityEnvironment.step": {
438
+ "total": 1810.2227486001043,
439
+ "count": 403195,
440
+ "is_parallel": true,
441
+ "self": 22.132137400104284,
442
+ "children": {
443
+ "UnityEnvironment._generate_step_input": {
444
+ "total": 22.734860300051785,
445
+ "count": 403195,
446
+ "is_parallel": true,
447
+ "self": 22.734860300051785
448
+ },
449
+ "communicator.exchange": {
450
+ "total": 1589.9010269999662,
451
+ "count": 403195,
452
+ "is_parallel": true,
453
+ "self": 1589.9010269999662
454
+ },
455
+ "steps_from_proto": {
456
+ "total": 175.45472389998184,
457
+ "count": 403195,
458
+ "is_parallel": true,
459
+ "self": 34.780302600093904,
460
+ "children": {
461
+ "_process_maybe_compressed_observation": {
462
+ "total": 125.5690601001191,
463
+ "count": 806390,
464
+ "is_parallel": true,
465
+ "self": 10.114038300169184,
466
+ "children": {
467
+ "_observation_to_np_array": {
468
+ "total": 115.45502179994992,
469
+ "count": 1210191,
470
+ "is_parallel": true,
471
+ "self": 9.955208899894956,
472
+ "children": {
473
+ "process_pixels": {
474
+ "total": 105.49981290005496,
475
+ "count": 1210191,
476
+ "is_parallel": true,
477
+ "self": 49.71157090014538,
478
+ "children": {
479
+ "image_decompress": {
480
+ "total": 55.78824199990958,
481
+ "count": 1210191,
482
+ "is_parallel": true,
483
+ "self": 55.78824199990958
484
+ }
485
+ }
486
+ }
487
+ }
488
+ }
489
+ }
490
+ },
491
+ "_process_rank_one_or_two_observation": {
492
+ "total": 15.105361199768835,
493
+ "count": 806390,
494
+ "is_parallel": true,
495
+ "self": 15.105361199768835
496
+ }
497
+ }
498
+ }
499
+ }
500
+ }
501
+ }
502
+ }
503
+ }
504
+ }
505
+ }
506
+ }
507
+ }
508
+ },
509
+ "trainer_threads": {
510
+ "total": 3.04000004689442e-05,
511
+ "count": 1,
512
+ "self": 3.04000004689442e-05,
513
+ "children": {
514
+ "thread_root": {
515
+ "total": 0.0,
516
+ "count": 0,
517
+ "is_parallel": true,
518
+ "self": 0.0,
519
+ "children": {
520
+ "trainer_advance": {
521
+ "total": 4283.552069900078,
522
+ "count": 208577,
523
+ "is_parallel": true,
524
+ "self": 4.947188199941593,
525
+ "children": {
526
+ "process_trajectory": {
527
+ "total": 3470.770895900137,
528
+ "count": 208577,
529
+ "is_parallel": true,
530
+ "self": 3470.3206993001368,
531
+ "children": {
532
+ "RLTrainer._checkpoint": {
533
+ "total": 0.4501966000000266,
534
+ "count": 2,
535
+ "is_parallel": true,
536
+ "self": 0.4501966000000266
537
+ }
538
+ }
539
+ },
540
+ "_update_policy": {
541
+ "total": 807.8339857999997,
542
+ "count": 138,
543
+ "is_parallel": true,
544
+ "self": 541.1189284,
545
+ "children": {
546
+ "TorchPPOOptimizer.update": {
547
+ "total": 266.7150573999998,
548
+ "count": 3354,
549
+ "is_parallel": true,
550
+ "self": 266.7150573999998
551
+ }
552
+ }
553
+ }
554
+ }
555
+ }
556
+ }
557
+ }
558
+ }
559
+ },
560
+ "TrainerController._save_models": {
561
+ "total": 0.24319540000033157,
562
+ "count": 1,
563
+ "self": 0.007341900000028545,
564
+ "children": {
565
+ "RLTrainer._checkpoint": {
566
+ "total": 0.23585350000030303,
567
+ "count": 1,
568
+ "self": 0.23585350000030303
569
+ }
570
+ }
571
+ }
572
+ }
573
+ }
574
+ }
575
+ }
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_7_task_3_run_id_2_train\\Agent\\Agent-499998.onnx",
13
+ "reward": 67.21430333851215,
14
+ "creation_time": 1717703680.289742,
15
+ "auxillary_file_paths": [
16
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent\\Agent-499998.pt"
17
+ ]
18
+ },
19
+ {
20
+ "steps": 999997,
21
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent\\Agent-999997.onnx",
22
+ "reward": 104.79983610265396,
23
+ "creation_time": 1717705600.2970777,
24
+ "auxillary_file_paths": [
25
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent\\Agent-999997.pt"
26
+ ]
27
+ },
28
+ {
29
+ "steps": 1201044,
30
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent\\Agent-1201044.onnx",
31
+ "reward": 97.25954774684377,
32
+ "creation_time": 1717706411.2942226,
33
+ "auxillary_file_paths": [
34
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent\\Agent-1201044.pt"
35
+ ]
36
+ }
37
+ ],
38
+ "final_checkpoint": {
39
+ "steps": 1201044,
40
+ "file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent.onnx",
41
+ "reward": 97.25954774684377,
42
+ "creation_time": 1717706411.2942226,
43
+ "auxillary_file_paths": [
44
+ "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_7_task_3_run_id_2_train\\Agent\\Agent-1201044.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
+ }