Gurobi Optimization Gurobi v4.5.1
GurobiOptimizationGurobiv4.5.1英文正式版(智慧引擎提供了新一代的高精確性的描繪軟體)
破解說明:
關掉主程式,破解檔放置於crack夾內,請將破解檔複製於主程式的安裝目錄內既可破解
內容說明:
Gurobi智慧引擎提供了新一代的高精確性的描繪方案。
英文說明:
TheGurobiOptimizerisastate-of-the-artsolver
forlinearprogramming(LP),quadraticprogramming
(QP)andmixed-integerprogramming(MILPandMIQP).
Itwasdesignedfromthegrounduptoexploitmodern
multi-coreprocessors.EveryGurobilicenseallows
parallelprocessing,andtheGurobiParallel
Optimizerisdeterministic:twoseparaterunsonthe
samemodelwillproduceidenticalsolutionpaths.

ForsolvingLPandQPmodels,theGurobiOptimizer
includeshigh-performanceimplementationsofthe
primalsimplexmethod,thedualsimplexmethod,and
aparallelbarriersolver.ForMILPandMIQPmodels,
theGurobiOptimizerincorporatesthelatestmethods
includingcuttingplanesandpowerfulsolution
heuristics.Allmodelsbenefitfromadvanced
presolvemethodstosimplifymodelsandslashsolve
times.

TheGurobiOptimizeriswritteninCandis
accessiblefromseverallanguages.Inadditiontoa
powerful,interactivePythoninterfaceanda
matrix-orientedCinterface,weprovide
object-orientedinterfacesfromC++,Java,Python,
andthe.NETlanguages.Theseinterfaceshaveall
beendesignedtobelightweightandeasytouse,
withthegoalofgreatlyenhancingtheaccessibility
ofourproducts.Andsincetheinterfacesare
lightweight,theyarefasteranduselessmemory
thanotherstandardinterfaces.Ouronline
documentation(QuickStartGuide,ExampleTourand
ReferenceManual)describestheuseofthese
interfaces.

Gurobiisalsoavailablethroughseveralpowerful
third-partymodelingsystemsincludingAIMMS,AMPL,
FRONTLINESOLVERS,GAMS,MPL,OptimJandTOMLAB.

Mostofthechangesinthe4.5releaseoftheGurobi
Optimizerarerelatedtoperformance.Usersof
previousversionswilltypicallynotneedtomake
anychangestotheirprogramstousethenew
version.Thenewversiondoescontainafewnew
features,describedhere.

*NewdefaultMethodforcontinuousmodels:The
newversionusesanewAutomaticsettingasthe
defaultforsolvingcontinuousmodels.Inprevious
releases,continuousmodelsweresolvedwiththe
dualsimplexmethodbydefault.Whiletheexact
strategyusedbythenewAutomaticsettingmay
changeinfuturereleases,inthisreleasethenew
approachusestheconcurrentoptimizerfor
continuousmodelswithalinearobjective(LPs),
thebarrieroptimizerforcontinuousmodelswitha
quadraticobjective(QPs),andthedualsimplex
optimizerfortherootnodeofaMIPmodel.You
shouldchangetheMethodparameterifyouwould
liketochooseadifferentmethod.

*NewMinimumReleaxationheuristic:Thenew
versioncontainsanewMinimumRelaxation
heuristicthatcanbeusefulforfindingsolutions
toMIPmodelswhereotherstrategiesfailtofind
feasiblesolutionsinareasonableamountoftime.
UsethenewMinRelNodesparametertocontrolthis
newheuristic.

*Newbranchdirectioncontrol:Thenewversion
allowsmorecontroloverhowthebranch-and-cut
treeisexplored.Specifically,whenanodeinthe
MIPsearchiscompletedandtwochildnodes,
correspondingtothedownbranchandtheupbranch
arecreated,thenewBranchDirparameterallows
youtodeterminewhethertheMIPsolverwill
explorethedownbranchfirst,theupbranch
first,orwhetheritwillchoosethenextnode
basedonaheuristicdeterminationofwhich
sub-treeappearsmorepromising.

*Cutpasslimit:Thenewversionallowsyouto
limitthenumberofcutpassesperformedduring
rootcutgenerationinMIP.UsethenewCutPasses
parameter.

*Additionalinformationforinfeasibleand
unboundedlinearmodels:Thenewversionallows
youtoobtainaFarkasinfeasibilityprooffor
infeasiblemodels,andanunboundedrayfor
unboundedmodels.UsethenewInfUnbdInfo
parameter,andthenewFarkasProof,FarkasDual,
UnbdRayattributestoobtainthisinformation.
圖片說明:

相關商品:WileyRFIDfortheOptimizationofBusinessProcessesApr2008英文正式版(PDF格式電子書)