README.md

---
tags:
- HalfCheetah-v4
- deep-reinforcement-learning
- reinforcement-learning
- custom-implementation
library_name: cleanrl
model-index:
- name: TD3
  results:
  - task:
      type: reinforcement-learning
      name: reinforcement-learning
    dataset:
      name: HalfCheetah-v4
      type: HalfCheetah-v4
    metrics:
    - type: mean_reward
      value: 8838.14 +/- 133.03
      name: mean_reward
      verified: false
---

# (CleanRL) **TD3** Agent Playing **HalfCheetah-v4**

This is a trained model of a TD3 agent playing HalfCheetah-v4.
The model was trained by using [CleanRL](https://github.com/vwxyzjn/cleanrl) and the most up-to-date training code can be
found [here](https://github.com/vwxyzjn/cleanrl/blob/master/cleanrl/td3_continuous_action_jax.py).

## Get Started

To use this model, please install the `cleanrl` package with the following command:

```
pip install "cleanrl[td3_continuous_action_jax]"
python -m cleanrl_utils.enjoy --exp-name td3_continuous_action_jax --env-id HalfCheetah-v4
```

Please refer to the [documentation](https://docs.cleanrl.dev/get-started/zoo/) for more detail.


## Command to reproduce the training

```bash
curl -OL https://huggingface.co/cleanrl/HalfCheetah-v4-td3_continuous_action_jax-seed1/raw/main/td3_continuous_action_jax.py
curl -OL https://huggingface.co/cleanrl/HalfCheetah-v4-td3_continuous_action_jax-seed1/raw/main/pyproject.toml
curl -OL https://huggingface.co/cleanrl/HalfCheetah-v4-td3_continuous_action_jax-seed1/raw/main/poetry.lock
poetry install --all-extras
python td3_continuous_action_jax.py --track --capture-video --env-id HalfCheetah-v4 --seed 1 --save-model --upload-model --hf-entity cleanrl
```

# Hyperparameters
```python
{'batch_size': 256,
 'buffer_size': 1000000,
 'capture_video': True,
 'env_id': 'HalfCheetah-v4',
 'exp_name': 'td3_continuous_action_jax',
 'exploration_noise': 0.1,
 'gamma': 0.99,
 'hf_entity': 'cleanrl',
 'learning_rate': 0.0003,
 'learning_starts': 25000.0,
 'noise_clip': 0.5,
 'policy_frequency': 2,
 'policy_noise': 0.2,
 'save_model': True,
 'seed': 1,
 'tau': 0.005,
 'total_timesteps': 1000000,
 'track': True,
 'upload_model': True,
 'wandb_entity': None,
 'wandb_project_name': 'cleanRL'}
```