AI4CPS OSS

AI4CPS is a Python framework for building ML & AI apps in the domain of Cyber-Physical Systems (CPSs).

Built on top of Dash, AI4CPS allows simple implementation of AI tools for CPSs. Read our tutorial.

This software is developed at Helmut Schmidt University / University of Federal Armed Forces Hamburg at the Professorship of Computer Science in Mechanical Engineering at the Institute of Automation Technology.

Installation

pip install ai4cps

The default installation does not require MLflow or PyTorch. It requires ml4cps>=0.2.13, where both dependencies are optional. For the MLflow-backed machine_learning() saved-model selector, install:

pip install "ai4cps[mlflow]"

The dashboard and feature execution work without MLflow. The dev extra includes MLflow for development and testing.

AI4CPS App Examples

To be added.

Install ai4cps and import from ai4cps, for example from ai4cps.dash.dashboard import Dash4CPS. The deprecated selfx distribution provides compatibility for legacy selfx imports.