This manual describes a new alternative to the Xamine histogram visualizer. It based on Python bindings for C++ and adopts common Python libraries for basic and more advanced numerical algorithms. The GUI is written on PyQt5. Among the new features, the ability to implement new Machine Learning (ML) algorithms for clustering and functions to fit spectra.
This document is organized as follows:
Description of CutiePie design.
Analysis features.
How to customize and add your own ML algorithms and fitting functions.
CutiePie is a histogram visualizer linked to the SpecTcl TreeGui. It accesses the shared memory location and imports spectum and gate definitions. The communication between CutiePie and SpecTcl is bidirectional. Creation of gates in CutiePie are mirrored in SpecTcl and viceversa. On the other hands, CutiePie extends some functionalities and provides hooks to Python-based analysis tools (i.e. Jupiter notebook).