欢迎来到天天文库
浏览记录
ID:34891221
大小:8.67 MB
页数:370页
时间:2019-03-13
《High Performance Python.pdf》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、HighPerformancePythonPRACTICALPERFORMANTPROGRAMMINGFORHUMANSMichaGorelick&IanOzsvaldwww.it-ebooks.infoHighPerformancePythonHighPerformancePythonYourPythoncodemayruncorrectly,butyouneedittorunfaster.Byexploringthefundamentaltheorybehinddesignchoices,thispractical“Despiteitspopularitygu
2、idehelpsyougainadeeperunderstandingofPython’simplementation.inacademiaandYou’lllearnhowtolocateperformancebottlenecksandsignificantlyspeedHighPerformanceindustry,Pythonisoftenupyourcodeinhigh-data-volumeprograms.dismissedastooslowforHowcanyoutakeadvantageofmulti-corearchitecturesorclu
3、sters?Orbuildasystemthatcanscaleupanddownwithoutlosingreliability?realapplications.ThisExperiencedPythonprogrammerswilllearnconcretesolutionstothesebooksweepsawaythatandotherissues,alongwithwarstoriesfromcompaniesthatusehighmisconceptionwithaperformancePythonforsocialmediaanalytics,pr
4、oductionizedmachinethoroughintroductionlearning,andothersituations.tostrategiesforfastandPython■Getabettergraspofnumpy,Cython,andprofilersscalablecomputation■LearnhowPythonabstractstheunderlyingcomputerwithPython.”architecture—JakeVanderPlasUniversityofWashingtonPRACTICALPERFORMANT■Us
5、eprofilingtofindbottlenecksinCPUtimeandmemoryusagePROGRAMMINGFORHUMANS■Writeefficientprogramsbychoosingappropriatedatastructures■Speedupmatrixandvectorcomputations■UsetoolstocompilePythondowntomachinecode■ManagemultipleI/Oandcomputationaloperationsconcurrently■Convertmultiprocessingco
6、detorunonalocalorremotecluster■SolvelargeproblemswhileusinglessRAMMichaGorelick,winneroftheNobelPrizein2046forhiscontributionstotimetravel,wentbacktothe2000stostudyastrophysics,workondataatbitly,andco-foundFastForwardLabsasresidentMadScientist,workingonissuesfrommachinelearningtoperfo
7、rmantstreamalgorithms.&OzsvaldIanOzsvaldisadatascientistandteacheratModelInsight.io,withovertenyearsGorelickofPythonexperience.He’staughthighperformancePythonatthePyConandPyDataconferencesandhasbeenconsultingondatascienceandhighperfor-mancecomputingforyearsintheUK.PYTHON/PERFORMANCETw
8、itter
此文档下载收益归作者所有