Examples#

This section provides practical examples demonstrating NeurEco Dynamic models using test cases included with the NeurEco installation package.

Test Cases Overview#

NeurEco includes several test cases for learning and experimentation:

Test Case

Template

Description

Nonlinear Oscillator

ODE

Duffing oscillator with strong nonlinearity

Mechanical TestBed

Convolutional Dynamic

Displacement of a mechanical test bench driven by three forces at a varying temperature

Other Available Test Cases#

The NeurEco installation includes additional test cases:

For ODE models:

  • Temperature Forecasting: Air temperature from concurrent weather-station measurements

  • Mass Spring Damper: Classic mechanical system (parametric inputs)

  • Electric Motor Temperature: Thermal modeling of electric motors

For Convolutional Dynamic models:

  • Air Pollution: Environmental time-series prediction

These test cases come with NeurEco as zip files (see Getting Started below).

What You’ll Learn#

These examples demonstrate:

  • Data loading: How to organize and load time-series data

  • Building: Configuring and training ODE and Convolutional Dynamic models

  • Evaluation: Computing errors and comparing predictions to reference data

  • Visualization: Plotting results for analysis

  • Best practices: Per-feature normalization, steady-state constraints, window size selection

Getting Started#

To run these examples:

  1. Find the test case’s zip file (for example NonLinearOscillator.zip). On Windows it is in <install folder>\testsuite\ (by default C:\Program Files\Adagos\NeurEco-5.0\testsuite\); on Linux, in /opt/neureco/share/adagos/.neurecotestsuite/. The pip package does not include the examples (see Examples)

  2. Extract the zip file to a working directory, or open the example from the desktop app’s example gallery

  3. Copy the example code and modify paths as needed

  4. Run the script to build and evaluate the model