Reinforcement Learning
ml-agents
TensorBoard
ONNX
Pyramids
deep-reinforcement-learning
ML-Agents-Pyramids
Instructions to use AnnaMats/ppo-Pyramids-Training with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ml-agents
How to use AnnaMats/ppo-Pyramids-Training with ml-agents:
mlagents-load-from-hf --repo-id="AnnaMats/ppo-Pyramids-Training" --local-dir="./download: string[]s"
- Notebooks
- Google Colab
- Kaggle
| using UnityEngine; | |
| public class PushBlockSettings : MonoBehaviour | |
| { | |
| /// <summary> | |
| /// The "walking speed" of the agents in the scene. | |
| /// </summary> | |
| public float agentRunSpeed; | |
| /// <summary> | |
| /// The agent rotation speed. | |
| /// Every agent will use this setting. | |
| /// </summary> | |
| public float agentRotationSpeed; | |
| /// <summary> | |
| /// The spawn area margin multiplier. | |
| /// ex: .9 means 90% of spawn area will be used. | |
| /// .1 margin will be left (so players don't spawn off of the edge). | |
| /// The higher this value, the longer training time required. | |
| /// </summary> | |
| public float spawnAreaMarginMultiplier; | |
| /// <summary> | |
| /// When a goal is scored the ground will switch to this | |
| /// material for a few seconds. | |
| /// </summary> | |
| public Material goalScoredMaterial; | |
| /// <summary> | |
| /// When an agent fails, the ground will turn this material for a few seconds. | |
| /// </summary> | |
| public Material failMaterial; | |
| } | |