Welcome to SPARK: Spectroscopy Pump-Probe Analysis and Research Kit
SPARK is an open-source Python suite designed for the processing and analysis of ultrafast spectroscopy data. It provides a comprehensive environment for handling transient absorption (TAS) and fluorescence up-conversion (FLUPS) measurements, featuring an integrated graphical user interface and a robust computational backend.
While it supports standard pre-processing tasks such as chirp correction and map generation, SPARK’s primary analytical focus is its Global Fitting Engine. This module implements Variable Projection (VarPro) algorithms to handle parallel and sequential kinetic models, as well as user-defined systems through custom rate matrices.
Core Capabilities
Global Fitting (VarPro): Separation of non-linear parameters (lifetimes, IRF width, etc.) from linear amplitude spectra using Trust Region Reflective optimization.
Kinetic Models: Support for Decay Associated Spectra (DAS), Species Associated Spectra (SAS), damped oscillations, and custom K-matrix models.
Statistical Diagnostics: Confidence interval estimation via profile likelihood mapping and condition number evaluation.
Data Pre-processing: Automated and manual data importers, baseline correction, and coherent artifact handling (Raman and XPM bases).
Publication Plotting: Integrated tools for customizing and exporting high-resolution vector figures.