Developing motion sensor apps with automatic trigger to Azure Machine Learning Studio Experiment to classify movement and visualise on a website
Repository includes
- database schema
- API
- example app
- example website
- function app code
- link to machine learning experiments
The app uses the following plugins:
- Plugin.BLE (version 1.3.0)
- Plugin.Movesense (1.1.0-beta)
- Microsoft.Azure.Devices (1.17.0)
For replication: The azure function trigger needs to be set to blob uploads to a storage account in the settings. Additionally an IoT hub needs to be configured, and connection strings need to be set up in the app.
Links to Azure Machine Learning Studio Experiments:
- Training and comparing different models: https://gallery.cortanaintelligence.com/Experiment/training-model
- Movement classification of incoming data using a trained model: https://gallery.cortanaintelligence.com/Experiment/sensor-evaluation-analysis