metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_3_task_2_run_id_0_train
tags:
- hivex
- hivex-aerial-wildfire-suppression
- reinforcement-learning
- multi-agent-reinforcement-learning
model-index:
- name: hivex-AWS-PPO-baseline-task-2-difficulty-3
results:
- task:
type: sub-task
name: maximize_preparing_non_burning_trees
task-id: 2
difficulty-id: 3
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.0833333358168602 +/- 0.1378617027533506
name: Crash Count
verified: true
- type: extinguishing_trees
value: 18.683333218097687 +/- 27.558120115325824
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 93.41666712760926 +/- 137.79060314424476
name: Extinguishing Trees Reward
verified: true
- type: fire_out
value: 0.35833333656191824 +/- 0.36784102365159543
name: Fire Out
verified: true
- type: fire_too_close_to_city
value: 0.975 +/- 0.11180339887498947
name: Fire too Close to City
verified: true
- type: preparing_trees
value: 556.9166661500931 +/- 491.4252421024537
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 2784.5833506584167 +/- 2457.1262108957085
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 39.591666674613954 +/- 20.687346400193867
name: Water Drop
verified: true
- type: water_pickup
value: 39.18333334922791 +/- 20.631967029054877
name: Water Pickup
verified: true
- type: cumulative_reward
value: 2911.9833114624025 +/- 2372.473518108635
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 3 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 2
Difficulty: 3
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment