Gadfly: A pandas-based Framework for Analyzing GADGET Simulation Data. (12th October 2016)

• Record Type: Journal Article
• Authors: Hummel, Jacob A.
• Is Part Of: Publications of the Astronomical Society of the Pacific. Volume 128:Number 969(2016)
• Publication Date: 2016-10-12
• Subjects: methods: data analysis -- methods: numerical; Astronomy -- Periodicals; Astronomy; Periodicals; Periodicals; 520.5
• DOI: 10.1088/1538-3873/128/969/114503
• Languages: English
• ISSNs: 0004-6280

Abstract

Abstract: We present the first public release (v0.1) of the open-sourcegadget Dataframe Library:gadfly . The aim of this package is to leverage the capabilities of the broader python scientific computing ecosystem by providing tools for analyzing simulation data from the astrophysical simulation codesgadget andgizmo usingpandas, a thoroughly documented, open-source library providing high-performance, easy-to-use data structures that is quickly becoming the standard for data analysis in python.Gadfly is a framework for analyzing particle-based simulation data stored in the HDF5 format usingpandas DataFrame s. The package enables efficient memory management, includes utilities for unit handling, coordinate transformations, and parallel batch processing, and provides highly optimized routines for visualizing smoothed-particle hydrodynamics data sets.