GAMSPy and Machine Learning

GAMSPy heralds a new era of possibilities, offering a bridge between machine learning and optimization that was previously challenging to traverse with GAMS alone. Here’s why GAMSPy stands out as the ultimate choice:

• Easy to understand, easy to write

  • eq1[...] = y == tanh(a @ W + b)

• Versatility in Solver Selection

  • regression.solve(solver="your_favourite_solver")

• It provides a strong algebraic language allows you to play with how neural network is implemented

• Built-in flexibility:

  • You are not limited with inference, you can try training your neural-network.

  • You can build the architecture from scratch in GAMSPy

• Development speed of Python combined with model generation speed of GAMS

  • Equations and variables are generated in GAMS not in Python giving GAMSPy a speed advantage

We are continuously developing our ML-related features. If you have specific needs or require additional information, please feel free to reach out to us at machinelearning@gams.com.

• Matrix Operations - Introduction
• Embedding a trained neural network in GAMSPy
• Regression Example
• Binary Classification Example
• Neural Network Example