DaphneKoller关于ProbabilisticGraphicalModels的最权威大作，内容详实深入，是各大名校机器学习和人工智能专业相应课程的指定教材AdaptiveComputationandMachineLearningThomasdietterich,EditorChristopherBishop,DavidHeckerman,MichaelJordan,andMichaelKearns,AssociateEditorsBioinformatics:TheMachinelearningApproach,PierreBaldiandSorenBrunakReinforcementLearning:AnIntroduction,RichardS.SuttonandAndrewG.BartoGraphicalmodelsforMachineLearningandDigitalCommunication,BrendanJ.FreyLearningingraphicalModels,MichaelI.JordanCausation,Prediction,andSearch,2nded,PeterSpirtes,ClarkGlymour,andRichardScheinesPrinciplesofDataMining,DavidHand,HeikkiMannila,andPadhraicSmythBioinformatics:TheMachineLearningApproach,2nded,PierreBaldiandSorenBrunakLearningKernelclassifiers:TheoryandAlgorithms,RalfHerbrichLearningwithKernels:SupportVectorMachines,Regularization,Optimization,andBeyond,BernhardScholkopfandAlexanderJsmolaIntroductiontoMachineLearning,EthemAlpaydinGaussianProcessesforMachineLearning,CarlEdwardRasmussenandChristopherK.I.WilliamsSemi-SupervisedLearning,OlivierChapelle,BernhardScholkopf,andAlexanderZien,edsTheMinimumdescriptionLengthPrinciple,PeterDGrunwaldIntroductiontoStatisticalRelationalLearning,liseGetoorandBenTaskar,edsProbabilisticGraphicalModels:PrinciplesandTechniques,DaphneKollerandNirFriedmanProbabilisticGraphicalModelsPrinciplesandTechniquesDaphnekollerNirfriedmanThemitpressCambridge,MassachusettsLondon,England@2009MassachusettsInstituteofTechnologyAllrightsreserved.Nopartofthisbookmaybereproducedinanyformbyanyelectronicormechanicalmeans(includingphotocopying,recording,orinformationstorageandretrieval)withoutpermissioninwritingfromthepublisherForinformationaboutspecialquantitydiscounts,pleaseemailspecial_sales@mitpress.mit.eduThisbookwassetbytheauthorsinBlFX2EPrintedandboundintheunitedstatesofamericaLibraryofCongressCataloging-in-PublicationDataKoller,DaphneProbabilisticGraphicalModels:PrinciplesandTechniquesDaphneKollerandNirFriedmanpcm.-(Adaptivecomputationandmachinelearning)IncludesbibliographicalreferencesandindexisBn978-0-262-01319-2(hardcover:alk.paper1.Graphicalmodeling(Statistics)2.Bayesianstatisticaldecisiontheory--Graphicmethods.IKoller,Daphne.II.Friedman,NirQA279.5.K652010519.5’420285-dc222009008615109876543ToourfamiliesmyparentsDovandditzamyhusbanddanmydaughtersnatalieandmayaDKmyparentsNogaandGadmywifemychildrenroyandliorMEAsfarasthelawsofmathematicsrefertoreality,theyarenotcertain,asfarastheyarecertain,theydonotrefertorealityAlberteinstein1956Whenwetrytopickoutanythingbyitself,wefindthatitisboundfastbyathousandinvisiblecordsthatcannotbebroken,toeverythingintheuniverseJohnMuir,1869Theactualscienceoflogicisconversantatpresentonlywiththingseithercertain,impossible,orentirelydoubtful.Thereforethetruelogicforthisworldisthecalculusofprobabilities,whichtakesaccountofthemagnitudeoftheprobabilitywhichis,oroughttobe,inareasonableman'smindJamesClerkMaxwell,1850Thetheoryofprobabilitiesisatbottomnothingbutcommonsensereducedtocalculus;itenablesustoappreciatewithexactnessthatwhichaccuratemindsfeelwithasortofinstinctforwhichofttimestheyareunabletoaccount.PierreSimonLaplace,1819MisunderstandingofprobabilitymaybethegreatestofallimpedimentstoscientificliteracyStephenJayGouldContentsAcknowledgmentsListoffiguresListofalgorithmsListofboxesXXX1IntroductionL1Motivation11.2StructuredProbabilisticModels21.2.1ProbabilisticGraphicalModels31.2.2Representation,Inference,Learning51.3Overviewandroadmap61.3.1OverviewofChapters61.3.2Readersguide1.3.3ConnectiontoOtherDisciplines1.4Historicalnotes122Foundations2.1ProbabilityTheory2.1.1ProbabilityDistributions152.1.2BasicConceptsinProbability182.1.3RandomVariablesandJointDistributions192.1.4IndependenceandConditionalIndependence2:2.1.5QueryingaDistribution2.1.6ContinuousSpaces272.1.7ExpectationandVariance312.2Graphs342.2.1Nodesandedges342.2.2Subgraphs352.2.3Pathsandtrails36

LSTM-Neural-Network-for-Time-Series-Prediction-master.rar

ThisisasmalllibrarythatcantrainRestrictedBoltzmannMachines,andalsoDeepBeliefNetworksofstackedRBM's.TrainRBM's:%trainanRBMwithbinaryvisibleunitsand500binaryhiddenmodel=rbmBB(data,500);%visualizethelearnedweightsvisualize(model.W);Doclassification:model=rbmFit(data,500,labels);prediction=rbmPredict(model,testdata);TrainaDeepBeliefNetworkwith500,500,2000architectureforclassification:models=dbnFit(data,[5005002000],labels);prediction=dbnPredict(models,testdata);seeincludedexamplecodeformoreIcanbecontactedonandrej.karpathy@gmail.NOTE:ThiswasaclassprojectthatIworkedonfor1monthandthenabandoneddevelopmentforalmost4yearsago.Pleasedonotsendmespecificquestionsaboutissueswiththecodeorquestionsonhowtodosomething.Ionlyputthiscodeonlineinhopethatitcanbeusefultoothersbutcannotfullysupportit.Ifyouwouldlikepointerstomoreactivelymaintainedimplementations,havealookhere(https://github.com/rasmusbergpalm/DeepLearnToolbox)ormaybehere(https://github.com/lisa-lab/DeepLearningTutorials)Sorryandbestofluck!原文：http://code.google.com/p/matrbm/

kaggle比赛数据集：Porto_Seguro’s_Safe_Driver_Prediction_all.zipPorto_Seguro’s_Safe_Driver_Prediction_all.zip

Loan_prediction:在DataHack上做预尝试-AnalyticsVidhya

罗斯曼销售预测注意：我主要使用PLOTLY在此笔记本中可视化我的数据，但可打印的图未显示在github上。
Kaggle竞赛：Rossmann在欧洲7个国家/地区拥有3,000多家药店。
目前，Rossmann商店经理的任务是提前六周预测其每日销售额。
商店的销售受到许多因素的影响，包括促销，竞争，学校和州假日，季节性和地区性。
成千上万的个人经理根据他们的独特情况预测销售额，结果的准确性可能会大相径庭。
我们接受Rossmann的挑战，即要预测整个德国1,115家商店的6周日销售量。
可靠的销售预测使商店经理能够创建有效的员工时间表，从而提高生产力和动力。
通过协助Rossmann创建可靠的预测模型，我们将协助商店经理专注于对他们而言最重要的事情：他们的顾客和他们的团队！目录进口包资料准备2.1加载数据集2.2处理缺失的值2.2.1计算每个数据集中的缺失值2.2.2删

在日常工作中，钉钉打卡成了我生活中不可或缺的一部分。然而，有时候这个看似简单的任务却给我带来了不少烦恼。 每天早晚，我总是得牢记打开钉钉应用，点击"工作台"，再找到"考勤打卡"进行签到。有时候因为工作忙碌，会忘记打卡，导致考勤异常，影响当月的工作评价。而且，由于我使用的是苹果手机，有时候系统更新后，钉钉的某些功能会出现异常，使得打卡变得更加麻烦。 另外，我的家人使用的是安卓手机，他们也经常抱怨钉钉打卡的繁琐。尤其是对于那些不太熟悉手机操作的长辈来说，每次打卡都是一次挑战。他们总是担心自己会操作失误，导致打卡失败。 为了解决这些烦恼，我开始思考是否可以通过编写一个全自动化脚本来实现钉钉打卡。经过一段时间的摸索和学习，我终于成功编写出了一个适用于苹果和安卓系统的钉钉打卡脚本。