metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_10_task_6_run_id_1_train
tags:
- hivex
- hivex-aerial-wildfire-suppression
- reinforcement-learning
- multi-agent-reinforcement-learning
model-index:
- name: hivex-AWS-PPO-baseline-task-6-difficulty-10
results:
- task:
type: sub-task
name: drop_water
task-id: 6
difficulty-id: 10
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.01919534709304571 +/- 0.004891916336268155
name: Crash Count
verified: true
- type: extinguishing_trees
value: 0.14774187933653593 +/- 0.21025496427030596
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 0.7387093845754862 +/- 1.0512748048411527
name: Extinguishing Trees Reward
verified: true
- type: preparing_trees
value: 275.08778228759763 +/- 6.872679334348032
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 275.08778228759763 +/- 6.872679334348032
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 0.9804799735546113 +/- 0.0052643568095604555
name: Water Drop
verified: true
- type: water_pickup
value: 0.0006513423752039671 +/- 0.0012320225884513893
name: Water Pickup
verified: true
- type: cumulative_reward
value: 273.9627319335938 +/- 7.442108906238094
name: Cumulative Reward
verified: true
This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 6 with difficulty 10 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 6
Difficulty: 10
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment