This repository is a machine learning library to deploy for Autoware to aim for "robotics MLOps".
Autoware-ml supports training with T4dataset format in addition to open datasets and deployment for Autoware.
In addition to ML model deployment, autoware-ml supports active learning framework include auto labeling, semi-auto labeling, and data mining.
Autoware-ml can deploy following task for now.
- 2D detection for dynamic recognition
- 3D detection for dynamic recognition
- 2D fine detection for traffic light recognition
- 2D classification for traffic light recognition
Autoware-ml supports following environment.
- All tools are tested by Docker environment on Ubuntu 22.04LTS
- NVIDIA dependency: CUDA 12.1 + cuDNN 8
- Need > 530.xx.xx NVIDIA device driver
If you want to know about the design of autoware-ml, you should read following pages.
- Docs for architecture of dataset pipeline
- Docs for architecture of ML model
- Docs for architecture of S3 storage
- Docs for autoware-ml design
If you want to develop autoware-ml, you should read following pages.
If you want to search the OSS tools around autoware-ml, you should read following pages.
If you want to know about autoware-ml, you should read following pages.
- Setting environment for autoware-ml
- Training and evaluation
- Analyze for the dataset and the model
- Auto labeling
- Data mining
- ROS2
- Model for Autoware
- Model for ML tools
- Model for Autoware
- Model for Autoware
- Model for ML tools
- Model for ML tools
- (TBD) SegmentAnything
- Model for Autoware
- Model for ML tools