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.

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:

pip install standard-e2e

Or for development:

git clone https://github.com/stepankonev/StandardE2E.git
cd StandardE2E
uv sync --all-extras

Refer to the Quickstart guide for detailed usage.

Documentation

Getting Started

• Quickstart
  • Installation
  • Preprocessing (required before training)
  • Interactive Tutorials
  • Code Examples
  • Adding New Datasets
  • Next Steps
  • Need Help?
• Overview
  • Why StandardE2E?
  • Architecture
  • Preprocessing
  • Training
• Supported Datasets
  • How datasets are added

API Reference

• API Reference
  • Unified Dataset
  • Data Structures
  • Enums
  • Caching and Processing
  • Dataset Utilities
  • Modality Defaults
  • Augmentation
  • Indexing

Tutorials

• 📓 Introduction Tutorial
• 📓 Data Containers
• 📓 Multi-Dataset Training
• 📓 Custom Adapters

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

• Index

• Module Index

• Search Page