<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<site>mtc-m16b.sid.inpe.br 802</site>
		<holdercode>{isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}</holdercode>
		<identifier>x6e6X3pFwXQZ3DUS8rS5/Cys5B</identifier>
		<repository>cptec.inpe.br/walmeida/2004/06.24.14.34</repository>
		<lastupdate>2015:11.18.17.41.27 cptec.inpe.br/walmeida/2003/04.25.17.12 marciana</lastupdate>
		<metadatarepository>cptec.inpe.br/walmeida/2004/06.24.14.34.11</metadatarepository>
		<metadatalastupdate>2018:06.05.03.51.12 cptec.inpe.br/walmeida/2003/04.25.17.12 administrator {D 2002}</metadatalastupdate>
		<secondarykey>INPE--PRE/</secondarykey>
		<citationkey>VijaykumarStePreCamNow:2002:OpNeNe</citationkey>
		<title>Optimized Neural Network Code for Data Assimilation</title>
		<format>CD-ROM</format>
		<year>2002</year>
		<date>4-9 ago.</date>
		<numberoffiles>1</numberoffiles>
		<size>170 KiB</size>
		<author>Vijaykumar, Nandamudi Lankalapali,</author>
		<author>Stephanyl, Stephan,</author>
		<author>Preto, Airam Jonatas,</author>
		<author>Campos Velho, Haroldo Fraga de,</author>
		<author>Nowosad, Alexandre G.,</author>
		<group>LAC-INPE-MCT-BR</group>
		<group>DMD-INPE-MCT-BR</group>
		<affiliation>INPE-Sao Jose dos Campos-12227-010-SP-Brasil</affiliation>
		<e-mailaddress>fabia@cptec.inpe.br</e-mailaddress>
		<conferencename>Congresso Brasileiro de Meteorologia, 12.</conferencename>
		<conferencelocation>Foz do Iguacu</conferencelocation>
		<pages>3841-3849</pages>
		<booktitle>Anais</booktitle>
		<secondarytype>PRE CN</secondarytype>
		<tertiarytype>Artigos</tertiarytype>
		<organization>SBMET</organization>
		<transferableflag>1</transferableflag>
		<contenttype>External Contribution</contenttype>
		<abstract>The data assimilation process can be described as a procedure that uses observational data to improve the prediction made by an inaccurate mathematical modelo Recent1y, neural networks have been proposed as a new method for data assimilation. The Multilayer Perceptron network with backpropagation learning was chosen for this procedure. Neural networks are inherent1y a parallel procedure. This paper presents some strategies being used to achieve an optimized parallel code for the network training. Code optimizations include the use of either High Perfonnance Fortran directives or Message Passing Interface library calls. A neural network for Data Assimilation was trained based on both the physical models of the Lorenz and shallow water equations.</abstract>
		<area>MET</area>
		<language>en</language>
		<usergroup>administrator</usergroup>
		<usergroup>marciana</usergroup>
		<readergroup>administrator</readergroup>
		<readergroup>marciana</readergroup>
		<visibility>shown</visibility>
		<readpermission>allow from all</readpermission>
		<documentstage>not transferred</documentstage>
		<nexthigherunit>8JMKD3MGPCW/3ESGTTP</nexthigherunit>
		<hostcollection>cptec.inpe.br/walmeida/2003/04.25.17.12</hostcollection>
		<username>marciana</username>
		<lasthostcollection>cptec.inpe.br/walmeida/2003/04.25.17.12</lasthostcollection>
		<url>http://mtc-m16b.sid.inpe.br/rep-/cptec.inpe.br/walmeida/2004/06.24.14.34</url>
	</metadata>
</metadatalist>