Starter Projects

Background

A starter project contains everything you need to quickly get started with your AI project:

• A project and data description
• Example data
• A trained model (.h5 Tensorflow model)
• Example predictions from the model.
• Converted edge code, i.e. a neural network model and preprocessor translated to C code, ready for integration into your embedded firmware.

We have added starter projects that cover a wide range of use cases and industrial applications. The projects are either of type Classification (Supervised learning) or Anomaly detection (Semi-supervised learning). You can choose among the following starter projects:

• Classification

  • Empty Classification Project
  • Human Activity Recognition
  • Acconeer Radar Gesture
  • Texas Instruments Radar Gesture
  • Renesas Touchpad Letter Detection
  • Keyword Spotter
  • Indoor/outdoor Detector

• Anomaly detection

  • Empty Anomaly Detection Project
  • Rotating Machinery (and similar periodic processes)

Get Started

To launch a Starter project, click on the Cube Icon in the upper left corner and select a project from the list that appears.

Select a name and location for your project and choose whether you want to download the pre-recorded data or not.

For each of the projects, documentation can be found by double clicking the README file in the downloaded project.