{"id":19976101,"url":"https://github.com/canoksuzoglu1/rl-pco-atlantis-atari","last_synced_at":"2026-05-14T19:04:37.551Z","repository":{"id":256996246,"uuid":"857053246","full_name":"canoksuzoglu1/RL-PCO-Atlantis-Atari","owner":"canoksuzoglu1","description":"This project trains and evaluates a Proximal Policy Optimization (PPO) agent to play the Atari game Atlantis using Stable Baselines3. The agent is trained with a Convolutional Neural Network (CNN) policy and evaluated for its performance in the game. It includes scripts for training, evaluating, and real-time gameplay rendering.","archived":false,"fork":false,"pushed_at":"2024-09-13T19:24:19.000Z","size":29880,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-12T09:47:51.702Z","etag":null,"topics":["atari","atlantis","cnn","deep-learning","game-ai","gym","machine-learning","openai-gym","ppo","reinforcement-learning","rl-agents","stable-baselines3","tensorboard"],"latest_commit_sha":null,"homepage":"https://gymnasium.farama.org/environments/atari/atlantis/","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/canoksuzoglu1.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-09-13T17:54:13.000Z","updated_at":"2024-09-29T15:46:54.000Z","dependencies_parsed_at":null,"dependency_job_id":"7986a761-8da2-4e11-98a8-5d9bca03c10b","html_url":"https://github.com/canoksuzoglu1/RL-PCO-Atlantis-Atari","commit_stats":null,"previous_names":["canoksuzoglu1/rl-pco-atlantis-atari"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/canoksuzoglu1%2FRL-PCO-Atlantis-Atari","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/canoksuzoglu1%2FRL-PCO-Atlantis-Atari/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/canoksuzoglu1%2FRL-PCO-Atlantis-Atari/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/canoksuzoglu1%2FRL-PCO-Atlantis-Atari/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/canoksuzoglu1","download_url":"https://codeload.github.com/canoksuzoglu1/RL-PCO-Atlantis-Atari/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":241411474,"owners_count":19958746,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["atari","atlantis","cnn","deep-learning","game-ai","gym","machine-learning","openai-gym","ppo","reinforcement-learning","rl-agents","stable-baselines3","tensorboard"],"created_at":"2024-11-13T03:21:53.945Z","updated_at":"2026-05-14T19:04:32.509Z","avatar_url":"https://github.com/canoksuzoglu1.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# RL-PCO-Atlantis-Atari\n\nThis project involves training and evaluating a Proximal Policy Optimization (PPO) agent to play the Atari game **Atlantis** using Stable Baselines3. The agent is trained with a Convolutional Neural Network (CNN) policy and evaluated for its performance in the game.\n\n## Project Overview\n\n1. **Library Imports**: Import necessary libraries including Gym and Stable Baselines3.\n2. **Environment Setup**: Configure the Atari environment and stack frames.\n3. **Model Initialization**: Initialize the PPO model with a CNN policy and configure logging.\n4. **Training the Model**: Train the PPO model for 2 million timesteps.\n5. **Model Saving and Loading**: Save the trained model and reload it for evaluation.\n6. **Evaluation and Rendering**: Evaluate the trained model and render the environment to visualize performance.\n7. **Real-Time Interaction**: Use the trained model to interact with the environment in real-time.\n\n## Requirements\n\n- Python 3.7+\n- Gym\n- Stable Baselines3\n- TensorBoard\n- Other dependencies can be installed via `requirements.txt`.\n\n\n## Project Structure\n\nThe project directory is organized as follows:\n\n- **`Atlantis RL PCO .ipynb`**: Jupyter notebook containing the main code and experiments.\n- **`README.md`**: This file, providing an overview and instructions for the project.\n- **`requirements.txt`**: File listing the required Python packages.\n- **`Training/`**: Directory containing training logs and saved models.\n  - **`Logs/`**: Subdirectory for TensorBoard logs.\n  - **`Saved Models/`**: Subdirectory for storing trained models.\n    - **`PPO_Atlantis_1MT.zip`**: Trained model saved after 1 million timesteps.\n    - **`PPO_Atlantis_2MT.zip`**: Trained model saved after 2 million timesteps.\n\n## Usage\n\n1. **Jupyter Notebook**: Open and run the Jupyter notebook `Atlantis RL PCO .ipynb` to explore and execute the main code and experiments. You can interactively train, evaluate, and visualize the PPO agent's performance within the notebook.\n\n2. **Training the Model**: Inside the Jupyter notebook, you can find cells for training the PPO model. Execute these cells to train the model with the desired configurations. \n\n3. **Evaluating the Model**: Similarly, use the evaluation cells in the notebook to assess the trained model's performance. You can visualize the results and analyze how well the model performs on the Atari Atlantis game.\n\n4. **Real-Time Playback**: To run the trained model in real-time and see the agent playing the game, you can add code within the notebook to load a saved model and render the gameplay.\n\n5. **Model Files**: After training, the model files will be saved in the `Training/Saved Models/` directory. You can use these files for later evaluation or real-time playback. The available models are:\n    - **`PPO_Atlantis_1MT.zip`**: Model trained for 1 million timesteps.\n    - **`PPO_Atlantis_2MT.zip`**: Model trained for 2 million timesteps.\n\n6. **Dependencies**: Install the required Python packages using the `requirements.txt` file:\n    ```bash\n    pip install -r requirements.txt\n    ```\n\n\n## License\n\nThis project is licensed under the MIT License.\n\n## Acknowledgements\n\n- [Stable Baselines3](https://stable-baselines3.readthedocs.io/en/master/)\n- [OpenAI Gym](https://www.gymlibrary.ml/)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcanoksuzoglu1%2Frl-pco-atlantis-atari","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcanoksuzoglu1%2Frl-pco-atlantis-atari","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcanoksuzoglu1%2Frl-pco-atlantis-atari/lists"}