Downloads · 30 days
41
30% of all-time downloads
PPO-SnowballTarget
kinkpunk/PPO-SnowballTarget
PPO-SnowballTarget is a reinforcement learning model from kinkpunk. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for ml-agents.
This is a trained model of a ppo agent playing SnowballTarget using the Unity ML-Agents Library.
All-time downloads
137
Public
Repo size
30.9 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt32.6 MB · 85%
At a glance
- Task
- Reinforcement Learning
- Library
- ml-agents
- Access
- Public
- Created
- Jan 17, 2023
- Updated
- Jan 17, 2023
- SHA
- a4364c5e
Model card
From the Hugging Face model README
ppo Agent playing SnowballTarget
This is a trained model of a ppo agent playing SnowballTarget using the Unity ML-Agents Library.
Usage (with ML-Agents)
The Documentation: https://github.com/huggingface/ml-agents#get-started We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
Resume the training
mlagents-learn ./config/ppo/SnowballTarget.yaml --run-id="SnowballTarget-v1" --resume
Training hyperparameters
behaviors:
SnowballTarget:
trainer_type: ppo
summary_freq: 10000
keep_checkpoints: 5
checkpoint_interval: 50000
max_steps: 900000
time_horizon: 128
threaded: true
hyperparameters:
learning_rate: 0.0001
learning_rate_schedule: linear
batch_size: 128
buffer_size: 4096
beta: 0.005
epsilon: 0.2
lambd: 0.95
num_epoch: 5
network_settings:
normalize: false
hidden_units: 256
num_layers: 3
vis_encode_type: simple
reward_signals:
extrinsic:
gamma: 0.99
strength: 1.0
Watch your Agent play
You can watch your agent playing directly in your browser:.
- Go to https://huggingface.co/spaces/unity/ML-Agents-SnowballTarget
- Step 1: Write your model_id: kinkpunk/PPO-SnowballTarget
- Step 2: Select your .nn /.onnx file
- Click on Watch the agent play 👀
- Task
- Reinforcement Learning
- Library
- ml-agents
- Created
- Jan 17, 2023
- Updated
- Jan 17, 2023