Examples

Short, self-contained scripts showing how to use MadSpace from Python. Each example uses only the pieces of the API it needs to make its point; see the Python API for the full reference.

Interfacing with MadSpace is done through NumPy, unless the example involves machine learning or a GPU, in which case it uses PyTorch instead. The code shown on these pages is taken verbatim from the .rst source and run as part of the test suite, so it is guaranteed to work against the current MadSpace build.

• Constructing and calling a phase-space mapping
• Defining fiducial cuts
• Building a mapping for a Feynman diagram
• Multiple on-shell configurations and integration order
• Calling a MadGraph-generated matrix element
• Running on a GPU
• The built-in PDF interpolator
• Building a simple integrator
• Neural importance sampling with MadNIS
• Fusing RNG, flow, mapping and matrix element into one graph