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) ·10.1萬字

A+H股雙重審計(jì)管制取消的經(jīng)濟(jì)后果:基于權(quán)益成本和審計(jì)質(zhì)量的研究
會(huì)員

作為一項(xiàng)在促進(jìn)會(huì)計(jì)師事務(wù)所增強(qiáng)獨(dú)立性和積累經(jīng)驗(yàn)、提高上市公司治理水平和會(huì)計(jì)信息質(zhì)量等方面均具有重要作用的審計(jì)制度,雙重審計(jì)制度及其變遷一直是會(huì)計(jì)與審計(jì)理論界、實(shí)務(wù)界十分關(guān)注的話題。然而,現(xiàn)有文獻(xiàn)基本局限于研究雙重審計(jì)管制取消導(dǎo)致的審計(jì)方面的后果,未充分關(guān)注該政策變化對(duì)資本市場的潛在影響;此外,現(xiàn)存關(guān)于雙重審計(jì)管制取消的審計(jì)后果的研究缺乏對(duì)引起公司放棄雙重審計(jì)后審計(jì)質(zhì)量下降的內(nèi)在機(jī)制的深入探索。對(duì)此,本書基于審計(jì)需求、審計(jì)管制、審計(jì)供給以及遵循效應(yīng)相關(guān)理論,采用我國上市公司數(shù)據(jù),借助經(jīng)典模型測算權(quán)益成本,新建指標(biāo)反映驅(qū)動(dòng)審計(jì)質(zhì)量變化的內(nèi)在機(jī)制,運(yùn)用多元回歸分析、傾向得分匹配、變化分析、混雜變量影響閾值分析等多種方法,對(duì)雙重審計(jì)管制取消在資本市場和審計(jì)市場上產(chǎn)生的經(jīng)濟(jì)后果進(jìn)行理論闡釋與實(shí)證檢驗(yàn)。本書為理解雙重審計(jì)制度及其變遷的影響提供新的參考,對(duì)我國今后的審計(jì)制度安排與行業(yè)發(fā)展具有重要啟示。

張睿 ·統(tǒng)計(jì) ·12萬字

QQ閱讀手機(jī)版

主站蜘蛛池模板: 福安市| 南皮县| 平顺县| 昌江| 闵行区| 仙居县| 新余市| 洛川县| 元江| 罗源县| 隆子县| 翁牛特旗| 平武县| 垣曲县| 金门县| 突泉县| 彩票| 克什克腾旗| 新密市| 白水县| 巴彦淖尔市| 南城县| 青阳县| 桂东县| 五河县| 化隆| 乌拉特中旗| 建宁县| 沁水县| 斗六市| 治多县| 浠水县| 手游| 扶沟县| 开化县| 汤阴县| 松原市| 曲周县| 浪卡子县| 湘阴县| 靖宇县|