metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_4_task_1_run_id_2_train
tags:
- hivex
- hivex-aerial-wildfire-suppression
- reinforcement-learning
- multi-agent-reinforcement-learning
model-index:
- name: hivex-AWS-PPO-baseline-task-1-difficulty-4
results:
- task:
type: sub-task
name: maximize_extinguished_burning_trees
task-id: 1
difficulty-id: 4
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.00833333358168602 +/- 0.03726780073566347
name: Crash Count
verified: true
- type: extinguishing_trees
value: 15.225000095367431 +/- 25.73818226110129
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 761.2499908447265 +/- 1286.9091013360119
name: Extinguishing Trees Reward
verified: true
- type: fire_out
value: 0.36666666716337204 +/- 0.4244018838665434
name: Fire Out
verified: true
- type: fire_too_close_to_city
value: 0.925 +/- 0.24468024246479642
name: Fire too Close to City
verified: true
- type: preparing_trees
value: 651.3250035211444 +/- 627.0399351081593
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 651.3250035211444 +/- 627.0399351081593
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 46.266666889190674 +/- 26.519958404149136
name: Water Drop
verified: true
- type: water_pickup
value: 45.76666655540466 +/- 26.436529035148055
name: Water Pickup
verified: true
- type: cumulative_reward
value: 1711.8633205413819 +/- 2423.7230528427467
name: Cumulative Reward
verified: true
This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 1 with difficulty 4 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 1
Difficulty: 4
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment