metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_7_task_2_run_id_2_train
tags:
- hivex
- hivex-aerial-wildfire-suppression
- reinforcement-learning
- multi-agent-reinforcement-learning
model-index:
- name: hivex-AWS-PPO-baseline-task-2-difficulty-7
results:
- task:
type: sub-task
name: maximize_preparing_non_burning_trees
task-id: 2
difficulty-id: 7
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.41666667610406877 +/- 0.29369595331135534
name: Crash Count
verified: true
- type: extinguishing_trees
value: 26.98611140549183 +/- 45.067904043919626
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 134.9305568575859 +/- 225.33952073669184
name: Extinguishing Trees Reward
verified: true
- type: fire_out
value: 0.09166666939854622 +/- 0.1375963286810061
name: Fire Out
verified: true
- type: fire_too_close_to_city
value: 0.9833333343267441 +/- 0.07453559480732508
name: Fire too Close to City
verified: true
- type: preparing_trees
value: 915.0361038208008 +/- 630.722632747104
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 4575.180584716797 +/- 3153.613200762862
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 63.81111078262329 +/- 27.460713193319517
name: Water Drop
verified: true
- type: water_pickup
value: 63.5777774810791 +/- 27.43349377033259
name: Water Pickup
verified: true
- type: cumulative_reward
value: 4956.189190673828 +/- 2576.604065716017
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 7 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 2
Difficulty: 7
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment