metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_1_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-1
results:
- task:
type: sub-task
name: drop_water
task-id: 6
difficulty-id: 1
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.040009206905961034 +/- 0.018735561549171307
name: Crash Count
verified: true
- type: extinguishing_trees
value: 0.4969854736700654 +/- 0.6300676451261423
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 2.484927378222346 +/- 3.150338277465211
name: Extinguishing Trees Reward
verified: true
- type: preparing_trees
value: 143.18572998046875 +/- 12.10208288767324
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 143.18572998046875 +/- 12.10208288767324
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 0.9589269667863846 +/- 0.018643650861373894
name: Water Drop
verified: true
- type: cumulative_reward
value: 141.81843147277831 +/- 13.9801382441756
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 1 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 6
Difficulty: 1
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment