Edit model card

(CleanRL) DQN Agent Playing BreakoutNoFrameskip-v4

This is a trained model of a DQN agent playing BreakoutNoFrameskip-v4. The model was trained by using CleanRL and the most up-to-date training code can be found here.

Command to reproduce the training

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

Hyperparameters

{'batch_size': 32,
 'buffer_size': 1000000,
 'capture_video': True,
 'end_e': 0.01,
 'env_id': 'BreakoutNoFrameskip-v4',
 'exp_name': 'dqn_atari_jax',
 'exploration_fraction': 0.1,
 'gamma': 0.99,
 'hf_entity': 'cleanrl',
 'learning_rate': 0.0001,
 'learning_starts': 80000,
 'save_model': True,
 'seed': 1,
 'start_e': 1,
 'target_network_frequency': 1000,
 'total_timesteps': 10000000,
 'track': True,
 'train_frequency': 4,
 'upload_model': True,
 'wandb_entity': None,
 'wandb_project_name': 'cleanRL'}
Downloads last month

-

Downloads are not tracked for this model. How to track
Video Preview
loading

Evaluation results

  • mean_reward on BreakoutNoFrameskip-v4
    self-reported
    291.10 +/- 116.43