Upload 13 files
Browse files- Agent.onnx +3 -0
- Agent/Agent-1201137.onnx +3 -0
- Agent/Agent-1201137.pt +3 -0
- Agent/Agent-499826.onnx +3 -0
- Agent/Agent-499826.pt +3 -0
- Agent/Agent-999942.onnx +3 -0
- Agent/Agent-999942.pt +3 -0
- Agent/checkpoint.pt +3 -0
- Agent/events.out.tfevents.1717248806.RICHARD.13176.0 +3 -0
- README.md +35 -0
- configuration.yaml +107 -0
- run_logs/timers.json +575 -0
- 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:ee45a939c383451f278b4bae0a4242d1d8e0f575948257a33043589631af9871
|
3 |
+
size 562656
|
Agent/Agent-1201137.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ee45a939c383451f278b4bae0a4242d1d8e0f575948257a33043589631af9871
|
3 |
+
size 562656
|
Agent/Agent-1201137.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:34fbefc56ce7720e6a9e46d92268f1a67a38c0988852b9402ab12da17eb6e8dc
|
3 |
+
size 4530031
|
Agent/Agent-499826.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cc493458c7b104cd498e1a1cbc634909b5f4e6f6e312a9c7733299767b953afd
|
3 |
+
size 562656
|
Agent/Agent-499826.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:337eda7cdd8e98b74ac205b75382be61f7bd23bee2090c3b799699cd17129aad
|
3 |
+
size 4530031
|
Agent/Agent-999942.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:14476be7f137ec5ae0f9f0f7eeb7519456c5445235a7565c11eb25b065294974
|
3 |
+
size 562656
|
Agent/Agent-999942.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0e246fb2536d52735926bb34ec042e1ae4823324223594a8c10148a97635fa7d
|
3 |
+
size 4530031
|
Agent/checkpoint.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:34fbefc56ce7720e6a9e46d92268f1a67a38c0988852b9402ab12da17eb6e8dc
|
3 |
+
size 4530031
|
Agent/events.out.tfevents.1717248806.RICHARD.13176.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3c90893a100abde313ee772423f9756874abd2aa6eb39b0ff6519f1f92420e32
|
3 |
+
size 2101926
|
README.md
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: hivex
|
3 |
+
original_train_name: DroneBasedReforestation_difficulty_4_task_2_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-2-difficulty-4
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
type: sub-task
|
14 |
+
name: pick_up_seed_at_base
|
15 |
+
task-id: 2
|
16 |
+
difficulty-id: 4
|
17 |
+
dataset:
|
18 |
+
name: hivex-drone-based-reforestation
|
19 |
+
type: hivex-drone-based-reforestation
|
20 |
+
metrics:
|
21 |
+
- type: out_of_energy_count
|
22 |
+
value: 0.5909523957967758 +/- 0.09171894105446358
|
23 |
+
name: Out of Energy Count
|
24 |
+
verified: true
|
25 |
+
- type: recharge_energy_count
|
26 |
+
value: 125.54469884961844 +/- 115.46428296295271
|
27 |
+
name: Recharge Energy Count
|
28 |
+
verified: true
|
29 |
+
- type: cumulative_reward
|
30 |
+
value: 12.542430520057678 +/- 7.328528013270426
|
31 |
+
name: Cumulative Reward
|
32 |
+
verified: true
|
33 |
+
---
|
34 |
+
|
35 |
+
This model serves as the baseline for the **Drone-Based Reforestation** environment, trained and tested on task <code>2</code> with difficulty <code>4</code> using the Proximal Policy Optimization (PPO) algorithm.<br><br>Environment: **Drone-Based Reforestation**<br>Task: <code>2</code><br>Difficulty: <code>4</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: 4
|
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: 2
|
94 |
+
name: task
|
95 |
+
completion_criteria: null
|
96 |
+
checkpoint_settings:
|
97 |
+
run_id: DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_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 +1,575 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"name": "root",
|
3 |
+
"gauges": {
|
4 |
+
"Agent.Policy.Entropy.mean": {
|
5 |
+
"value": 1.4389417171478271,
|
6 |
+
"min": 1.4189385175704956,
|
7 |
+
"max": 1.442765235900879,
|
8 |
+
"count": 200
|
9 |
+
},
|
10 |
+
"Agent.Policy.Entropy.sum": {
|
11 |
+
"value": 9781.92578125,
|
12 |
+
"min": 7058.5087890625,
|
13 |
+
"max": 9982.1494140625,
|
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": 82.5,
|
30 |
+
"min": 0.0,
|
31 |
+
"max": 494.1333333333333,
|
32 |
+
"count": 200
|
33 |
+
},
|
34 |
+
"Agent.DroneBasedReforestation.RechargeEnergyCount.sum": {
|
35 |
+
"value": 1980.0,
|
36 |
+
"min": 0.0,
|
37 |
+
"max": 7412.0,
|
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": 0.5833333333333334,
|
54 |
+
"min": 0.38095238095238093,
|
55 |
+
"max": 0.7333333333333333,
|
56 |
+
"count": 200
|
57 |
+
},
|
58 |
+
"Agent.DroneBasedReforestation.OutofEnergyCount.sum": {
|
59 |
+
"value": 14.0,
|
60 |
+
"min": 6.0,
|
61 |
+
"max": 15.0,
|
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.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": 283.0,
|
150 |
+
"min": 250.75,
|
151 |
+
"max": 399.0,
|
152 |
+
"count": 200
|
153 |
+
},
|
154 |
+
"Agent.Environment.EpisodeLength.sum": {
|
155 |
+
"value": 6792.0,
|
156 |
+
"min": 4884.0,
|
157 |
+
"max": 6954.0,
|
158 |
+
"count": 200
|
159 |
+
},
|
160 |
+
"Agent.Step.mean": {
|
161 |
+
"value": 1199937.0,
|
162 |
+
"min": 5600.0,
|
163 |
+
"max": 1199937.0,
|
164 |
+
"count": 200
|
165 |
+
},
|
166 |
+
"Agent.Step.sum": {
|
167 |
+
"value": 1199937.0,
|
168 |
+
"min": 5600.0,
|
169 |
+
"max": 1199937.0,
|
170 |
+
"count": 200
|
171 |
+
},
|
172 |
+
"Agent.Policy.CuriosityValueEstimate.mean": {
|
173 |
+
"value": 0.2653595805168152,
|
174 |
+
"min": 0.029307620599865913,
|
175 |
+
"max": 0.8288605213165283,
|
176 |
+
"count": 200
|
177 |
+
},
|
178 |
+
"Agent.Policy.CuriosityValueEstimate.sum": {
|
179 |
+
"value": 5.8379106521606445,
|
180 |
+
"min": 0.43961429595947266,
|
181 |
+
"max": 14.840982437133789,
|
182 |
+
"count": 200
|
183 |
+
},
|
184 |
+
"Agent.Policy.ExtrinsicValueEstimate.mean": {
|
185 |
+
"value": 2.1637535095214844,
|
186 |
+
"min": 0.07653423398733139,
|
187 |
+
"max": 3.4586215019226074,
|
188 |
+
"count": 200
|
189 |
+
},
|
190 |
+
"Agent.Policy.ExtrinsicValueEstimate.sum": {
|
191 |
+
"value": 47.602577209472656,
|
192 |
+
"min": 1.071479320526123,
|
193 |
+
"max": 64.29899597167969,
|
194 |
+
"count": 200
|
195 |
+
},
|
196 |
+
"Agent.Environment.CumulativeReward.mean": {
|
197 |
+
"value": 17.454433063976467,
|
198 |
+
"min": -1.0649333159128824,
|
199 |
+
"max": 38.686707599957785,
|
200 |
+
"count": 200
|
201 |
+
},
|
202 |
+
"Agent.Environment.CumulativeReward.sum": {
|
203 |
+
"value": 383.99752740748227,
|
204 |
+
"min": -15.973999738693237,
|
205 |
+
"max": 582.9760599876754,
|
206 |
+
"count": 200
|
207 |
+
},
|
208 |
+
"Agent.Policy.CuriosityReward.mean": {
|
209 |
+
"value": 0.746854851488024,
|
210 |
+
"min": 0.0,
|
211 |
+
"max": 14.208348274230957,
|
212 |
+
"count": 200
|
213 |
+
},
|
214 |
+
"Agent.Policy.CuriosityReward.sum": {
|
215 |
+
"value": 16.430806732736528,
|
216 |
+
"min": 0.0,
|
217 |
+
"max": 255.75026893615723,
|
218 |
+
"count": 200
|
219 |
+
},
|
220 |
+
"Agent.Policy.ExtrinsicReward.mean": {
|
221 |
+
"value": 15.708987276408482,
|
222 |
+
"min": -0.9584401766459147,
|
223 |
+
"max": 34.81802397171656,
|
224 |
+
"count": 200
|
225 |
+
},
|
226 |
+
"Agent.Policy.ExtrinsicReward.sum": {
|
227 |
+
"value": 345.5977200809866,
|
228 |
+
"min": -14.37660264968872,
|
229 |
+
"max": 524.6783403318841,
|
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": 0.02009747641083474,
|
246 |
+
"min": 0.015691468220514555,
|
247 |
+
"max": 0.034741795350176595,
|
248 |
+
"count": 139
|
249 |
+
},
|
250 |
+
"Agent.Losses.PolicyLoss.sum": {
|
251 |
+
"value": 0.02009747641083474,
|
252 |
+
"min": 0.015691468220514555,
|
253 |
+
"max": 0.034741795350176595,
|
254 |
+
"count": 139
|
255 |
+
},
|
256 |
+
"Agent.Losses.ValueLoss.mean": {
|
257 |
+
"value": 7.238462497790654,
|
258 |
+
"min": 0.009945196118981888,
|
259 |
+
"max": 26.402251879374187,
|
260 |
+
"count": 139
|
261 |
+
},
|
262 |
+
"Agent.Losses.ValueLoss.sum": {
|
263 |
+
"value": 7.238462497790654,
|
264 |
+
"min": 0.009945196118981888,
|
265 |
+
"max": 26.402251879374187,
|
266 |
+
"count": 139
|
267 |
+
},
|
268 |
+
"Agent.Policy.LearningRate.mean": {
|
269 |
+
"value": 1.8758493747499982e-06,
|
270 |
+
"min": 1.8758493747499982e-06,
|
271 |
+
"max": 0.0002979000007,
|
272 |
+
"count": 139
|
273 |
+
},
|
274 |
+
"Agent.Policy.LearningRate.sum": {
|
275 |
+
"value": 1.8758493747499982e-06,
|
276 |
+
"min": 1.8758493747499982e-06,
|
277 |
+
"max": 0.0002979000007,
|
278 |
+
"count": 139
|
279 |
+
},
|
280 |
+
"Agent.Policy.Epsilon.mean": {
|
281 |
+
"value": 0.10062525,
|
282 |
+
"min": 0.10062525,
|
283 |
+
"max": 0.1993,
|
284 |
+
"count": 139
|
285 |
+
},
|
286 |
+
"Agent.Policy.Epsilon.sum": {
|
287 |
+
"value": 0.10062525,
|
288 |
+
"min": 0.10062525,
|
289 |
+
"max": 0.1993,
|
290 |
+
"count": 139
|
291 |
+
},
|
292 |
+
"Agent.Policy.Beta.mean": {
|
293 |
+
"value": 4.119997499999998e-05,
|
294 |
+
"min": 4.119997499999998e-05,
|
295 |
+
"max": 0.00496507,
|
296 |
+
"count": 139
|
297 |
+
},
|
298 |
+
"Agent.Policy.Beta.sum": {
|
299 |
+
"value": 4.119997499999998e-05,
|
300 |
+
"min": 4.119997499999998e-05,
|
301 |
+
"max": 0.00496507,
|
302 |
+
"count": 139
|
303 |
+
},
|
304 |
+
"Agent.Losses.CuriosityForwardLoss.mean": {
|
305 |
+
"value": 0.02619731449522078,
|
306 |
+
"min": 0.024310729233548045,
|
307 |
+
"max": 0.5835290277997652,
|
308 |
+
"count": 139
|
309 |
+
},
|
310 |
+
"Agent.Losses.CuriosityForwardLoss.sum": {
|
311 |
+
"value": 0.02619731449522078,
|
312 |
+
"min": 0.024310729233548045,
|
313 |
+
"max": 0.5835290277997652,
|
314 |
+
"count": 139
|
315 |
+
},
|
316 |
+
"Agent.Losses.CuriosityInverseLoss.mean": {
|
317 |
+
"value": 1.9983123789230983,
|
318 |
+
"min": 1.9722897013028462,
|
319 |
+
"max": 3.310828596353531,
|
320 |
+
"count": 139
|
321 |
+
},
|
322 |
+
"Agent.Losses.CuriosityInverseLoss.sum": {
|
323 |
+
"value": 1.9983123789230983,
|
324 |
+
"min": 1.9722897013028462,
|
325 |
+
"max": 3.310828596353531,
|
326 |
+
"count": 139
|
327 |
+
}
|
328 |
+
},
|
329 |
+
"metadata": {
|
330 |
+
"timer_format_version": "0.1.0",
|
331 |
+
"start_time_seconds": "1717248804",
|
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_4_task_2_run_id_1_train.yaml --run-id=DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_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": "1717252409"
|
340 |
+
},
|
341 |
+
"total": 3604.5230556,
|
342 |
+
"count": 1,
|
343 |
+
"self": 0.3121977000005245,
|
344 |
+
"children": {
|
345 |
+
"run_training.setup": {
|
346 |
+
"total": 0.04929309999999998,
|
347 |
+
"count": 1,
|
348 |
+
"self": 0.04929309999999998
|
349 |
+
},
|
350 |
+
"TrainerController.start_learning": {
|
351 |
+
"total": 3604.1615647999997,
|
352 |
+
"count": 1,
|
353 |
+
"self": 5.150592599950414,
|
354 |
+
"children": {
|
355 |
+
"TrainerController._reset_env": {
|
356 |
+
"total": 2.1376168,
|
357 |
+
"count": 1,
|
358 |
+
"self": 2.1376168
|
359 |
+
},
|
360 |
+
"TrainerController.advance": {
|
361 |
+
"total": 3596.69150620005,
|
362 |
+
"count": 401217,
|
363 |
+
"self": 4.888665899957232,
|
364 |
+
"children": {
|
365 |
+
"env_step": {
|
366 |
+
"total": 3591.8028403000926,
|
367 |
+
"count": 401217,
|
368 |
+
"self": 1598.6335483001171,
|
369 |
+
"children": {
|
370 |
+
"SubprocessEnvManager._take_step": {
|
371 |
+
"total": 1989.9212179998997,
|
372 |
+
"count": 401217,
|
373 |
+
"self": 10.38291419988036,
|
374 |
+
"children": {
|
375 |
+
"TorchPolicy.evaluate": {
|
376 |
+
"total": 1979.5383038000193,
|
377 |
+
"count": 400379,
|
378 |
+
"self": 1979.5383038000193
|
379 |
+
}
|
380 |
+
}
|
381 |
+
},
|
382 |
+
"workers": {
|
383 |
+
"total": 3.2480740000757544,
|
384 |
+
"count": 401217,
|
385 |
+
"self": 0.0,
|
386 |
+
"children": {
|
387 |
+
"worker_root": {
|
388 |
+
"total": 3595.269167799993,
|
389 |
+
"count": 401217,
|
390 |
+
"is_parallel": true,
|
391 |
+
"self": 2201.3683121999547,
|
392 |
+
"children": {
|
393 |
+
"steps_from_proto": {
|
394 |
+
"total": 0.006103399999999981,
|
395 |
+
"count": 1,
|
396 |
+
"is_parallel": true,
|
397 |
+
"self": 0.00010250000000011639,
|
398 |
+
"children": {
|
399 |
+
"_process_maybe_compressed_observation": {
|
400 |
+
"total": 0.005953199999999992,
|
401 |
+
"count": 2,
|
402 |
+
"is_parallel": true,
|
403 |
+
"self": 2.859999999982321e-05,
|
404 |
+
"children": {
|
405 |
+
"_observation_to_np_array": {
|
406 |
+
"total": 0.005924600000000169,
|
407 |
+
"count": 3,
|
408 |
+
"is_parallel": true,
|
409 |
+
"self": 2.8500000000208914e-05,
|
410 |
+
"children": {
|
411 |
+
"process_pixels": {
|
412 |
+
"total": 0.00589609999999996,
|
413 |
+
"count": 3,
|
414 |
+
"is_parallel": true,
|
415 |
+
"self": 0.00023630000000007811,
|
416 |
+
"children": {
|
417 |
+
"image_decompress": {
|
418 |
+
"total": 0.005659799999999882,
|
419 |
+
"count": 3,
|
420 |
+
"is_parallel": true,
|
421 |
+
"self": 0.005659799999999882
|
422 |
+
}
|
423 |
+
}
|
424 |
+
}
|
425 |
+
}
|
426 |
+
}
|
427 |
+
}
|
428 |
+
},
|
429 |
+
"_process_rank_one_or_two_observation": {
|
430 |
+
"total": 4.7699999999872844e-05,
|
431 |
+
"count": 2,
|
432 |
+
"is_parallel": true,
|
433 |
+
"self": 4.7699999999872844e-05
|
434 |
+
}
|
435 |
+
}
|
436 |
+
},
|
437 |
+
"UnityEnvironment.step": {
|
438 |
+
"total": 1393.8947522000383,
|
439 |
+
"count": 401217,
|
440 |
+
"is_parallel": true,
|
441 |
+
"self": 17.846898699817302,
|
442 |
+
"children": {
|
443 |
+
"UnityEnvironment._generate_step_input": {
|
444 |
+
"total": 19.743010899945794,
|
445 |
+
"count": 401217,
|
446 |
+
"is_parallel": true,
|
447 |
+
"self": 19.743010899945794
|
448 |
+
},
|
449 |
+
"communicator.exchange": {
|
450 |
+
"total": 1214.4236552000543,
|
451 |
+
"count": 401217,
|
452 |
+
"is_parallel": true,
|
453 |
+
"self": 1214.4236552000543
|
454 |
+
},
|
455 |
+
"steps_from_proto": {
|
456 |
+
"total": 141.88118740022088,
|
457 |
+
"count": 401217,
|
458 |
+
"is_parallel": true,
|
459 |
+
"self": 27.934730900210923,
|
460 |
+
"children": {
|
461 |
+
"_process_maybe_compressed_observation": {
|
462 |
+
"total": 101.46559579985372,
|
463 |
+
"count": 802434,
|
464 |
+
"is_parallel": true,
|
465 |
+
"self": 8.125860299694509,
|
466 |
+
"children": {
|
467 |
+
"_observation_to_np_array": {
|
468 |
+
"total": 93.33973550015921,
|
469 |
+
"count": 1204416,
|
470 |
+
"is_parallel": true,
|
471 |
+
"self": 7.935106600225282,
|
472 |
+
"children": {
|
473 |
+
"process_pixels": {
|
474 |
+
"total": 85.40462889993393,
|
475 |
+
"count": 1204416,
|
476 |
+
"is_parallel": true,
|
477 |
+
"self": 40.33854679993299,
|
478 |
+
"children": {
|
479 |
+
"image_decompress": {
|
480 |
+
"total": 45.06608210000093,
|
481 |
+
"count": 1204416,
|
482 |
+
"is_parallel": true,
|
483 |
+
"self": 45.06608210000093
|
484 |
+
}
|
485 |
+
}
|
486 |
+
}
|
487 |
+
}
|
488 |
+
}
|
489 |
+
}
|
490 |
+
},
|
491 |
+
"_process_rank_one_or_two_observation": {
|
492 |
+
"total": 12.480860700156235,
|
493 |
+
"count": 802434,
|
494 |
+
"is_parallel": true,
|
495 |
+
"self": 12.480860700156235
|
496 |
+
}
|
497 |
+
}
|
498 |
+
}
|
499 |
+
}
|
500 |
+
}
|
501 |
+
}
|
502 |
+
}
|
503 |
+
}
|
504 |
+
}
|
505 |
+
}
|
506 |
+
}
|
507 |
+
}
|
508 |
+
},
|
509 |
+
"trainer_threads": {
|
510 |
+
"total": 2.4199999643315095e-05,
|
511 |
+
"count": 1,
|
512 |
+
"self": 2.4199999643315095e-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": 3600.703408599978,
|
522 |
+
"count": 175653,
|
523 |
+
"is_parallel": true,
|
524 |
+
"self": 3.4899279999713144,
|
525 |
+
"children": {
|
526 |
+
"process_trajectory": {
|
527 |
+
"total": 2869.1280733000076,
|
528 |
+
"count": 175653,
|
529 |
+
"is_parallel": true,
|
530 |
+
"self": 2868.7198174000077,
|
531 |
+
"children": {
|
532 |
+
"RLTrainer._checkpoint": {
|
533 |
+
"total": 0.40825589999985823,
|
534 |
+
"count": 2,
|
535 |
+
"is_parallel": true,
|
536 |
+
"self": 0.40825589999985823
|
537 |
+
}
|
538 |
+
}
|
539 |
+
},
|
540 |
+
"_update_policy": {
|
541 |
+
"total": 728.0854072999994,
|
542 |
+
"count": 139,
|
543 |
+
"is_parallel": true,
|
544 |
+
"self": 488.24386740000165,
|
545 |
+
"children": {
|
546 |
+
"TorchPPOOptimizer.update": {
|
547 |
+
"total": 239.8415398999977,
|
548 |
+
"count": 3372,
|
549 |
+
"is_parallel": true,
|
550 |
+
"self": 239.8415398999977
|
551 |
+
}
|
552 |
+
}
|
553 |
+
}
|
554 |
+
}
|
555 |
+
}
|
556 |
+
}
|
557 |
+
}
|
558 |
+
}
|
559 |
+
},
|
560 |
+
"TrainerController._save_models": {
|
561 |
+
"total": 0.1818249999996624,
|
562 |
+
"count": 1,
|
563 |
+
"self": 0.0060155999995004095,
|
564 |
+
"children": {
|
565 |
+
"RLTrainer._checkpoint": {
|
566 |
+
"total": 0.17580940000016199,
|
567 |
+
"count": 1,
|
568 |
+
"self": 0.17580940000016199
|
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": 499826,
|
12 |
+
"file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-499826.onnx",
|
13 |
+
"reward": -0.9994999766349792,
|
14 |
+
"creation_time": 1717250298.3967845,
|
15 |
+
"auxillary_file_paths": [
|
16 |
+
"results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-499826.pt"
|
17 |
+
]
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"steps": 999942,
|
21 |
+
"file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-999942.onnx",
|
22 |
+
"reward": -0.998800003528595,
|
23 |
+
"creation_time": 1717251804.8955786,
|
24 |
+
"auxillary_file_paths": [
|
25 |
+
"results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-999942.pt"
|
26 |
+
]
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"steps": 1201137,
|
30 |
+
"file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-1201137.onnx",
|
31 |
+
"reward": 15.915000875356297,
|
32 |
+
"creation_time": 1717252409.069626,
|
33 |
+
"auxillary_file_paths": [
|
34 |
+
"results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-1201137.pt"
|
35 |
+
]
|
36 |
+
}
|
37 |
+
],
|
38 |
+
"final_checkpoint": {
|
39 |
+
"steps": 1201137,
|
40 |
+
"file_path": "results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent.onnx",
|
41 |
+
"reward": 15.915000875356297,
|
42 |
+
"creation_time": 1717252409.069626,
|
43 |
+
"auxillary_file_paths": [
|
44 |
+
"results\\DroneBasedReforestation/train/DroneBasedReforestation_difficulty_4_task_2_run_id_1_train\\Agent\\Agent-1201137.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 |
+
}
|