metadata
library_name: hivex
original_train_name: AerialWildfireSuppression_difficulty_3_task_6_run_id_2_train
tags:
- hivex
- hivex-aerial-wildfire-suppression
- reinforcement-learning
- multi-agent-reinforcement-learning
model-index:
- name: hivex-AWS-PPO-baseline-task-6-difficulty-3
results:
- task:
type: sub-task
name: drop_water
task-id: 6
difficulty-id: 3
dataset:
name: hivex-aerial-wildfire-suppression
type: hivex-aerial-wildfire-suppression
metrics:
- type: crash_count
value: 0.01590130275581032 +/- 0.009051027459662414
name: Crash Count
verified: true
- type: extinguishing_trees
value: 0.2118260153569281 +/- 0.3283880873127659
name: Extinguishing Trees
verified: true
- type: extinguishing_trees_reward
value: 1.0591301042586565 +/- 1.641940515491148
name: Extinguishing Trees Reward
verified: true
- type: preparing_trees
value: 234.83799896240234 +/- 10.68919451687246
name: Preparing Trees
verified: true
- type: preparing_trees_reward
value: 234.83799896240234 +/- 10.68919451687246
name: Preparing Trees Reward
verified: true
- type: water_drop
value: 0.9835368990898132 +/- 0.009798197368354513
name: Water Drop
verified: true
- type: water_pickup
value: 0.000875470065511763 +/- 0.001850762177686202
name: Water Pickup
verified: true
- type: cumulative_reward
value: 234.31187438964844 +/- 9.555234680707661
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 3 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 6
Difficulty: 3
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment