官术网_书友最值得收藏!

Learn Amazon SageMaker
會(huì)員

Quicklybuildanddeploymachinelearningmodelswithoutmanaginginfrastructure,andimproveproductivityusingAmazonSageMaker’scapabilitiessuchasAmazonSageMakerStudio,Autopilot,Experiments,Debugger,andModelMonitorKeyFeatures*Build,train,anddeploymachinelearningmodelsquicklyusingAmazonSageMaker*Analyze,detect,andreceivealertsrelatingtovariousbusinessproblemsusingmachinelearningalgorithmsandtechniques*Improveproductivitybytrainingandfine-tuningmachinelearningmodelsinproductionBookDescriptionAmazonSageMakerenablesyoutoquicklybuild,train,anddeploymachinelearning(ML)modelsatscale,withoutmanaginganyinfrastructure.IthelpsyoufocusontheMLproblemathandanddeployhigh-qualitymodelsbyremovingtheheavyliftingtypicallyinvolvedineachstepoftheMLprocess.ThisbookisacomprehensiveguidefordatascientistsandMLdeveloperswhowanttolearntheinsandoutsofAmazonSageMaker.You’llunderstandhowtousevariousmodulesofSageMakerasasingletoolsettosolvethechallengesfacedinML.Asyouprogress,you’llcoverfeaturessuchasAutoML,built-inalgorithmsandframeworks,andtheoptionforwritingyourowncodeandalgorithmstobuildMLmodels.Later,thebookwillshowyouhowtointegrateAmazonSageMakerwithpopulardeeplearninglibrariessuchasTensorFlowandPyTorchtoincreasethecapabilitiesofexistingmodels.You’llalsolearntogetthemodelstoproductionfasterwithminimumeffortandatalowercost.Finally,you’llexplorehowtouseAmazonSageMakerDebuggertoanalyze,detect,andhighlightproblemstounderstandthecurrentmodelstateandimprovemodelaccuracy.BytheendofthisAmazonbook,you’llbeabletouseAmazonSageMakeronthefullspectrumofMLworkflows,fromexperimentation,training,andmonitoringtoscaling,deployment,andautomation.Whatyouwilllearn*Createandautomateend-to-endmachinelearningworkflowsonAmazonWebServices(AWS)*Becomewell-versedwithdataannotationandpreparationtechniques*UseAutoMLfeaturestobuildandtrainmachinelearningmodelswithAutoPilot*Createmodelsusingbuilt-inalgorithmsandframeworksandyourowncode*TraincomputervisionandNLPmodelsusingreal-worldexamples*Covertrainingtechniquesforscaling,modeloptimization,modeldebugging,andcostoptimization*AutomatedeploymenttasksinavarietyofconfigurationsusingSDKandseveralautomationtoolsWhothisbookisforThisbookisforsoftwareengineers,machinelearningdevelopers,datascientists,andAWSuserswhoarenewtousingAmazonSageMakerandwanttobuildhigh-qualitymachinelearningmodelswithoutworryingaboutinfrastructure.KnowledgeofAWSbasicsisrequiredtograsptheconceptscoveredinthisbookmoreeffectively.SomeunderstandingofmachinelearningconceptsandthePythonprogramminglanguagewillalsobebeneficial.

Julien Simon;Francesco Pochetti ·統(tǒng)計(jì) ·10.1萬(wàn)字

非線性經(jīng)濟(jì)關(guān)系的建模
會(huì)員

美國(guó)聯(lián)邦儲(chǔ)備委員會(huì)和許多國(guó)家中央銀行都在使用的評(píng)估和預(yù)測(cè)方法。大數(shù)據(jù)時(shí)代,人們尊重?cái)?shù)據(jù),分析數(shù)據(jù),解釋數(shù)據(jù),才能把握和預(yù)測(cè)未來(lái)。分析和利用數(shù)據(jù)都需要模型建模,特別是非線性模型建模。因?yàn)榇蠖鄶?shù)經(jīng)濟(jì)變量具有非線性關(guān)系,經(jīng)濟(jì)是非線性的。時(shí)間序列模型的主要功能就是預(yù)測(cè),因此非線性時(shí)間序列計(jì)量經(jīng)濟(jì)學(xué)成為計(jì)量經(jīng)濟(jì)學(xué)拓展演化的必然結(jié)果。本書(shū)是計(jì)量經(jīng)濟(jì)學(xué)文獻(xiàn)中關(guān)于非線性時(shí)間序列計(jì)量經(jīng)濟(jì)學(xué)專著。它從正確性和實(shí)用性兩個(gè)方面判斷模型,主要關(guān)注便于計(jì)量經(jīng)濟(jì)學(xué)家使用的相對(duì)簡(jiǎn)單的模型。具有如下特點(diǎn):i)給出時(shí)間序列計(jì)量經(jīng)濟(jì)學(xué)中非線性的定義;ii)綜述常用的非線性時(shí)間序列計(jì)量經(jīng)濟(jì)模型及其經(jīng)濟(jì)理論依據(jù);iii)系統(tǒng)地提出了非線性時(shí)間序列計(jì)量經(jīng)濟(jì)模型建模的三步驟,即模型設(shè)定,參數(shù)估計(jì),建模評(píng)估;iv)系統(tǒng)論述了Granger的長(zhǎng)記憶模型;v)提供了大量實(shí)證研究示例。本書(shū)的出版對(duì)研究財(cái)富與消費(fèi)、匯率與價(jià)格、以及短期利率與長(zhǎng)期利率之間的關(guān)系等問(wèn)題都具有非常重要的意義。

(英)克萊夫·格蘭杰 (芬)蒂莫·泰雷斯維爾塔 ·統(tǒng)計(jì) ·8.6萬(wàn)字

農(nóng)業(yè)內(nèi)部審計(jì)那些事兒
會(huì)員

盡管我國(guó)有很長(zhǎng)的農(nóng)業(yè)史,但在農(nóng)業(yè)現(xiàn)代化上,我們?nèi)栽趭^力追趕,而我們所遇到的挑戰(zhàn)將越來(lái)越多。本書(shū)主要針對(duì)農(nóng)業(yè)企業(yè)內(nèi)部審計(jì)中會(huì)遇到的一些難題,作了一些經(jīng)驗(yàn)總結(jié),同時(shí)為從事農(nóng)業(yè)管理業(yè)務(wù)的人士提供一些參考資料,盡可能地避開(kāi)欺詐、舞弊、侵權(quán)、不當(dāng)?shù)美痊F(xiàn)象。目前,市面上有很多關(guān)于內(nèi)部審計(jì)的書(shū),但大多數(shù)是從審計(jì)職能這個(gè)角度進(jìn)行論述,缺乏企業(yè)內(nèi)審角度的著作,因此,本書(shū)是從農(nóng)企的角度進(jìn)行論述。很多人認(rèn)為內(nèi)部審計(jì)的方法是不分行業(yè)的,但筆者認(rèn)為,在實(shí)際工作中還是有差別的。例如,農(nóng)產(chǎn)品盤(pán)點(diǎn)比工業(yè)企業(yè)難,農(nóng)產(chǎn)品受苗種、病害、退化、天氣等方面影響大。農(nóng)業(yè)有其自身的經(jīng)營(yíng)模式,本書(shū)從農(nóng)企的特點(diǎn)出發(fā),輔以實(shí)戰(zhàn)和經(jīng)驗(yàn),選取一些經(jīng)常用到的審計(jì)程序和審計(jì)項(xiàng)目進(jìn)行分析,使讀者能夠真正掌握農(nóng)業(yè)內(nèi)部審計(jì)的精髓。本書(shū)內(nèi)容精練、重點(diǎn)突出、實(shí)例豐富,是廣大農(nóng)業(yè)內(nèi)部審計(jì)人員必備的參考書(shū),同時(shí)也非常適合大中專院校師生學(xué)習(xí)閱讀,也可供農(nóng)業(yè)相關(guān)管理者作為培訓(xùn)教材使用。

甘德東 ·統(tǒng)計(jì) ·7.7萬(wàn)字

QQ閱讀手機(jī)版

主站蜘蛛池模板: 昭通市| 武乡县| 当涂县| 渭南市| 呼和浩特市| 松原市| 南郑县| 巴塘县| 成都市| 哈巴河县| 九龙城区| 扶风县| 凤阳县| 连平县| 鹤山市| 会泽县| 黄骅市| 平定县| 石狮市| 鄂州市| 东辽县| 咸阳市| 大连市| 香格里拉县| 西和县| 台东县| 封丘县| 大港区| 旌德县| 石河子市| 汨罗市| 昌都县| 米脂县| 宁蒗| 苗栗县| 临夏市| 岚皋县| 柳州市| 永吉县| 红河县| 吉安县|