GNA: new framework for statistical data analysis

Abstract

We report on the status of GNA -a new framework for fitting largescale physical models. GNA utilizes the data flow concept within which a model is represented by a directed acyclic graph. Each node is an operation on an array (matrix multiplication, derivative or cross section calculation, etc). The framework enables the user to create flexible and efficient large-scale lazily evaluated models, handle large numbers of parameters, propagate parameters' uncertainties while taking into account possible correlations between them, fit models, and perform statistical analysis. The main goal of the paper is to give an overview of the main concepts and methods as well as reasons behind their design. Detailed technical information is to be published in further works. *

Grammar Parser-Based Solution for the Description of the Computational Graph for the Gna Framework

“…GNA is a framework for fitting large-scale physical models [1], for example a Daya Bay model [2]. The framework uses the data flow concept [3] within which a model is represented by a directed acyclic graph.…”

“…GNA expressions are the concept that uses mathematical expressions to describe the relationships between transformations and bindings [1]. An expression is initialized with the following information: a) a mathematical expression, b) the definition of indices used in the expression, c) configurations for the bundles to be executed to provide expression elements.…”

A semi-analytical energy response model for low-energy events in JUNO

“…Figure 1.Expected energy spectra of the prompt IBD candidates in the JUNO experiment in no oscillation (dashed line) and three-neutrino oscillation models assuming normal (solid blue line) and inverted (solid red line) MO. The calculation was performed with the GNA software[9] assuming 3%/ E prompt [MeV] energy resolution and normal ordering oscillation parameters from[10]. To display the inverted ordering, only the sign of ∆m 2 32 was inverted.…”