metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_8_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-8
results:
- task:
type: sub-task
name: drop_water
task-id: 6
difficulty-id: 8
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.010958191571990027 +/- 0.004573914836666436
name: Crash Count
verified: true
- type: extinguishing_trees
value: 0.18876909412210807 +/- 0.14536552612874615
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 0.9438454601913691 +/- 0.7268276219188166
name: Extinguishing Trees Reward
verified: true
- type: preparing_trees
value: 295.6805847167969 +/- 8.689367684892192
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 295.6805847167969 +/- 8.689367684892192
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 0.9887655645608902 +/- 0.004140661152640941
name: Water Drop
verified: true
- type: water_pickup
value: 0.0006430462468415499 +/- 0.0013853557027389734
name: Water Pickup
verified: true
- type: cumulative_reward
value: 295.6266799926758 +/- 9.439199947060839
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 8 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 6
Difficulty: 8
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment