To use the full functionality of the IoT²-Pilotyzer, the following software packages should be installed:
Python version 3.5 or higher
The current Python versions are available for download here. Additional packages as NumPy or Scipy (Python libraries which provide useful functions) can be installed via the command prompt with the package manager PIP ( e.g. ‚python -m pip install SciPy‘).Required packages are:- NumPy
- SciPy
- joblib
- pandas
- scikit-learn
Node RED
The installation of nodeJS is the first step, which is available here. Then node-RED can be installed with the command ’npm install -g -unsafe-perm node-red‘ in the command prompt. In Node-RED missing modules like the node-red-dashboard can be installed in the menu under Manage palette. Recommended modules are:- node-red-dashboard
- node-red-contrib-mqtt-broker
- node-red-contrib-machine-learning
IoT²-Pilotyzer
After the software IoT²-Pilotyzer is downloaded, unzip the file close to the HD, so that the path is very short e.g. (c:\iotw-pilotyzer). Afterwards the program can be executed with start_windows. A new window will be opened in the browser. Now it can be checked whether all necessary modules are installed, or it can be continued without checking. With ‘GENERATE DATA‘ Ardublockly is started, with ‘PROCESS DATA‘ Node-RED and ‘ANALYZE DATA‘ calls the Node-RED dashboard. For the deployment of a microcontroller with Ardublockly, the correct port can be selected under settings, in the menu (upper left corner).If no suitable port is displayed, it may be necessary to install the appropriate driver for the microcontroller used.