opendilab/DI-drive
An open-source simulation platform for training autonomous driving policies using reinforcement learning, imitation learning, and other decision intelligence methods.
Not currently ranked — collecting fresh signals.
star history
DI-drive provides a unified platform for running imitation learning, reinforcement learning, and GAIL algorithms in autonomous driving simulations. It integrates with driving simulators like CARLA and MetaDrive, and builds on the DI-engine reinforcement learning framework to train driving policies. The platform supports configurable scenarios and decision intelligence evaluation across different simulation environments.
Frequently asked
- What is opendilab/DI-drive?
- An open-source simulation platform for training autonomous driving policies using reinforcement learning, imitation learning, and other decision intelligence methods.
- Is DI-drive open source?
- Yes — opendilab/DI-drive is open source, released under the Apache-2.0 license.
- What language is DI-drive written in?
- opendilab/DI-drive is primarily written in Python.
- How popular is DI-drive?
- opendilab/DI-drive has 636 stars on GitHub.
- Where can I find DI-drive?
- opendilab/DI-drive is on GitHub at https://github.com/opendilab/DI-drive.