ppo-SnowballTarget / README.md
lambdavi's picture
Update README.md
216e3eb verified
|
raw
history blame
2.45 kB
---
library_name: ml-agents
tags:
- SnowballTarget
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-SnowballTarget
---
# **ppo** Agent playing **SnowballTarget**
This is a trained model of a **ppo** agent playing **SnowballTarget**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog ๐Ÿถ to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: lambdavi/ppo-SnowballTarget
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play ๐Ÿ‘€
### Hyperparams used:
'''
SnowballTarget:
trainer_type: ppo
hyperparameters:
batch_size: 128
buffer_size: 2048
learning_rate: 0.005
beta: 0.005
epsilon: 0.2
lambd: 0.95
num_epoch: 5
shared_critic: False
learning_rate_schedule: linear
beta_schedule: linear
epsilon_schedule: linear
checkpoint_interval: 50000
network_settings:
normalize: False
hidden_units: 256
num_layers: 2
vis_encode_type: simple
memory: None
goal_conditioning_type: hyper
deterministic: False
reward_signals:
extrinsic:
gamma: 0.99
strength: 1.0
network_settings:
normalize: False
hidden_units: 128
num_layers: 2
vis_encode_type: simple
memory: None
goal_conditioning_type: hyper
deterministic: False
init_path: None
keep_checkpoints: 10
even_checkpoints: False
max_steps: 500000
time_horizon: 64
summary_freq: 10000
threaded: True
self_play: None
behavioral_cloning: None
'''