ADF95: Tool for automatic differentiation of a FORTRAN code designed for large numbers of independent variables

Published: 1 June 2005| Version 1 | DOI: 10.17632/6f5837dx3z.1
Christian W. Straka


Abstract ADF95 is a tool to automatically calculate numerical first derivatives for any mathematical expression as a function of user defined independent variables. Accuracy of derivatives is achieved within machine precision. ADF95 may be applied to any FORTRAN 77/90/95 conforming code and requires minimal changes by the user. It provides a new derived data type that holds the value and derivatives and applies forward differencing by overloading all FORTRAN operators and intrinsic functions. An effic... Title of program: ADF95 Catalogue Id: ADVI_v1_0 Nature of problem In many areas in the computational sciences first order partial derivatives for large and complex sets of equations are needed with machine precision accuracy. For example, any implicit or semi-implicit solver requires the computation of the Jacobian matrix, which contains the first derivatives with respect to the independent variables. ADF95 is a software module to facilitate the automatic computation of the first partial derivatives of any arbitrarily complex mathematical FORTRAN expression. T ... Versions of this program held in the CPC repository in Mendeley Data ADVI_v1_0; ADF95; 10.1016/j.cpc.2005.01.011 This program has been imported from the CPC Program Library held at Queen's University Belfast (1969-2018)



Computational Physics, Computational Method