metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_5_task_2_run_id_1_train
tags:
- hivex
- hivex-aerial-wildfire-suppression
- reinforcement-learning
- multi-agent-reinforcement-learning
model-index:
- name: hivex-AWS-PPO-baseline-task-2-difficulty-5
results:
- task:
type: sub-task
name: maximize_preparing_non_burning_trees
task-id: 2
difficulty-id: 5
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.25000000521540644 +/- 0.26765169236445613
name: Crash Count
verified: true
- type: extinguishing_trees
value: 26.541666515916585 +/- 71.99835260789617
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 132.70833393633364 +/- 359.9917639236706
name: Extinguishing Trees Reward
verified: true
- type: fire_out
value: 0.1916666693985462 +/- 0.3301381975702533
name: Fire Out
verified: true
- type: fire_too_close_to_city
value: 0.975 +/- 0.11180339887498947
name: Fire too Close to City
verified: true
- type: preparing_trees
value: 884.999996471405 +/- 751.8528699660257
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 4425.0000263214115 +/- 3759.264413506219
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 53.89166679382324 +/- 27.40039192850323
name: Water Drop
verified: true
- type: water_pickup
value: 53.56666660308838 +/- 27.298244112627053
name: Water Pickup
verified: true
- type: cumulative_reward
value: 5174.121683979034 +/- 4124.406680944283
name: Cumulative Reward
verified: true
This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 2 with difficulty 5 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 2
Difficulty: 5
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment