Automatic differentiation in Odyssée
Automatic differentiation in Odyssée
复制标题
DOI：
10.3402/tellusa.v45i5.15060
复制
发表时间：
1993-10
期刊：
影响因子：
--
通讯作者：
Nicole Rostaing;Stéphane Dalmas;A. Galligo
中科院分区：
文献类型：
--
作者：
Nicole Rostaing;Stéphane Dalmas;A. Galligo
本文介绍了一个Fortran程序处理系统Odyssee的设计及其在自动微分中的应用。Odyssee系统将Fortran程序作为符号对象进行操作。它是一个开放的系统，作为一个工具包，用一种高级编程语言编写。使用变分方法进行资料同化需要计算由大型Fortran程序表示的代价函数的梯度。反向自动微分法的常见缺点是存储要求。Odyssee系统允许实现存储/重新计算策略，以适应所需的折衷。我们提出了在天气预报arpege/ifs项目中使用的策略的实现，以产生伴随代码的代码表示的数值模式。Odyssee生成的代码与arpege/ifs项目的手写伴随代码相同。DOI：10.1034/j.1600-0870.1993.00016.x
This paper describes the design of Odyssee , a system for fortran programs manipulations and its application to automatic differentiation. The Odyssee system manipulates fortran programs as symbolic objects. It is an open system built as a toolkit, written in a high-level programming language adapted to this purpose. The use of a variational method to perform data assimilation requires the computation of the gradient of a cost function represented by a large-size fortran program. The usual drawback in the reverse automatic differentiation method is the storage requirement. The Odyssee system allows one to implement storage/recomputation strategies in order to fit the needed compromizes. We present the implementation of the strategy used in the weather forecasting arpege/ifs project to produce the adjoint code from the code representing the numerical model. Odyssee produces the same code as the hand-written adjoint code for thearpege/ifs project. DOI: 10.1034/j.1600-0870.1993.00016.x