StandardE2E Documentation ========================= **A framework for unified end-to-end autonomous driving datasets processing** StandardE2E provides a consistent interface for preprocessing, loading, and training with multimodal data from various end-to-end autonomous driving datasets. It standardizes the complex process of working with different dataset formats, allowing researchers to focus on model development rather than data engineering. .. image:: ../assets/standard_e2e_scheme.png :alt: StandardE2E Architecture :align: center :width: 100% | Key Features ------------ ✨ **Unified Data Format and API** - Single representation and consistent interface for all datasets 🔄 **Multimodal Support** - Camera, LiDAR, HD maps, trajectories, detections, driving command, etc ⚙️ **Parametrizable Pipelines** - Configure programmatically or via YAML files 🚀 **PyTorch Native** - Seamless DataLoader integration 📦 **Extensible** - Add new datasets, adapters, augmentations, etc. Getting Started --------------- Install from PyPI: .. code-block:: bash pip install standard-e2e Or for development: .. code-block:: bash git clone https://github.com/stepankonev/StandardE2E.git cd StandardE2E uv sync --all-extras Refer to the :doc:`quickstart` guide for detailed usage. Documentation ------------- .. toctree:: :maxdepth: 2 :caption: Getting Started quickstart overview datasets .. toctree:: :maxdepth: 2 :caption: API Reference reference/api .. toctree:: :maxdepth: 1 :caption: Tutorials 📓 Introduction Tutorial 📓 Data Containers 📓 Multi-Dataset Training 📓 Custom Adapters .. toctree:: :caption: Project Links :hidden: GitHub Discord PyPI Community & Support ------------------- Reach out to the maintainers via GitHub issues or Discord. - **GitHub Issues**: https://github.com/stepankonev/StandardE2E/issues - **Discord**: https://discord.gg/vJnQNcQGQ8 Indices and tables ================== * :ref:`genindex` * :ref:`modindex` * :ref:`search`