Advanced Deep Learning with Keras, published by Packt

Please note that the code examples have been updated to support TensorFlow 2.0 Keras API only. About the Book. Advanced Deep Learning with TensorFlow 2 and ... Skiptocontent {{message}} PacktPublishing / Advanced-Deep-Learning-with-Keras Public Notifications Fork 797 Star 1.3k AdvancedDeepLearningwithKeras,publishedbyPackt License MITlicense 1.3k stars 797 forks Star Notifications Code Issues 1 Pullrequests 0 Actions Wiki Security Insights More Code Issues Pullrequests Actions Wiki Security Insights PacktPublishing/Advanced-Deep-Learning-with-Keras Thiscommitdoesnotbelongtoanybranchonthisrepository,andmaybelongtoaforkoutsideoftherepository. master Branches Tags Couldnotloadbranches Nothingtoshow {{refName}} default Couldnotloadtags Nothingtoshow {{refName}} default 1 branch 0 tags Code Latestcommit   Gitstats 441 commits Files Permalink Failedtoloadlatestcommitinformation. Type Name Latestcommitmessage Committime chapter1-keras-quick-tour optionalpydot Sep24,2020 chapter10-policy policygradfixes Apr13,2020 chapter11-detection fnfixes Nov16,2020 chapter12-segmentation maskdisplay Apr16,2020 chapter13-mi-unsupervised model.fit Apr19,2020 chapter2-deep-networks plot_model Sep24,2020 chapter3-autoencoders UpdateREADME.md Feb21,2020 chapter4-gan codemigrationtotf.keras Dec14,2019 chapter5-improved-gan refactor Jan25,2020 chapter6-disentangled-gan infoganmilossfix May6,2020 chapter7-cross-domain-gan tfweights Dec19,2019 chapter8-vae UpdateREADME.md Feb21,2020 chapter9-drl dqndoc Jan26,2020 lib fcneval Jan15,2020 LICENSE Initialcommit Mar15,2018 README.md remove$10campaign Jun14,2022 requirements.txt reqsfixes Apr8,2020 Viewcode AdvancedDeepLearningwithTensorFlow2andKeras(Updatedfor2ndEdition) AbouttheBook RelatedProducts Installation AlternativeTensorFlowInstallation AdvancedDeepLearningwithTensorFlow2andKerascodeexamplesusedinthebook. Chapter1-Introduction Chapter2-DeepNetworks Chapter3-AutoEncoders Chapter4-GenerativeAdversarialNetwork(GAN) Chapter5-ImprovedGAN Chapter6-GANwithDisentangledLatentRepresentations Chapter7-Cross-DomainGAN Chapter8-VariationalAutoencoders(VAE) Chapter9-DeepReinforcementLearning Chapter10-PolicyGradientMethods Chapter11-ObjectDetection Chapter12-SemanticSegmentation Chapter13-UnsupervisedLearningusingMutualInformation Citation README.md AdvancedDeepLearningwithTensorFlow2andKeras(Updatedfor2ndEdition) ThisisthecoderepositoryforAdvancedDeepLearningwithTensorFlow2andKeras,publishedbyPackt.Itcontainsallthesupportingprojectfilesnecessarytoworkthroughthebookfromstarttofinish. PleasenotethatthecodeexampleshavebeenupdatedtosupportTensorFlow2.0KerasAPIonly. AbouttheBook AdvancedDeepLearningwithTensorFlow2andKeras,SecondEditionisacompletelyupdatededitionofthebestsellingguidetotheadvanceddeeplearningtechniquesavailabletoday.RevisedforTensorFlow2.x,thiseditionintroducesyoutothepracticalsideofdeeplearningwithnewchaptersonunsupervisedlearningusingmutualinformation,objectdetection(SSD),andsemanticsegmentation(FCNandPSPNet),furtherallowingyoutocreateyourowncutting-edgeAIprojects. UsingKerasasanopen-sourcedeeplearninglibrary,thebookfeatureshands-onprojectsthatshowyouhowtocreatemoreeffectiveAIwiththemostup-to-datetechniques. Startingwithanoverviewofmulti-layerperceptrons(MLPs),convolutionalneuralnetworks(CNNs),andrecurrentneuralnetworks(RNNs),thebookthenintroducesmorecutting-edgetechniquesasyouexploredeepneuralnetworkarchitectures,includingResNetandDenseNet,andhowtocreateautoencoders.YouwillthenlearnaboutGANs,andhowtheycanunlocknewlevelsofAIperformance. Next,you’lldiscoverhowavariationalautoencoder(VAE)isimplemented,andhowGANsandVAEshavethegenerativepowertosynthesizedatathatcanbeextremelyconvincingtohumans.You'llalsolearntoimplementDRLsuchasDeepQ-LearningandPolicyGradientMethods,whicharecriticaltomanymodernresultsinAI. RelatedProducts DeepReinforcementLearningHands-On DeepLearningwithKeras ReinforcementLearningwithTensorFlow Installation Itisrecommendedtorunwithincondaenvironment.PlsdownloadAnacodafrom:Anaconda.Toinstallanaconda: sh Amachinewithatleast1NVIDIAGPU(1060orbetter)isrequired.Thecodeexampleshavebeentestedon1060,1080Ti,RTX2080Ti,V100,RTXQuadro8000onUbuntu18.04LTS.BelowisaroughguidetoinstallNVIDIAdriverandCuDNNtoenableGPUsupport. sudoadd-apt-repositoryppa:graphics-drivers/ppa sudoaptupdate sudoubuntu-driversautoinstall sudoreboot nvidia-smi Atthetimeofwriting,nvidia-smishowstheNVIDIAdriverversionis440.64andCUDAversionis10.2. Wearealmostthere.Thelastsetofpackagesmustbeinstalledasfollows.Somestepsmightrequiresudoaccess. condacreate--namepackt condaactivatepackt cd gitclonehttps://github.com/PacktPublishing/Advanced-Deep-Learning-with-Keras cdAdvanced-Deep-Learning-with-Keras pipinstall-rrequirements.txt sudoapt-getinstallpython-pydot sudoapt-getinstallffmpeg Testifasimplemodelcanbetrainedwithouterrors: cdchapter1-keras-quick-tour python3mlp-mnist-1.3.2.py ThefinaloutputshowstheaccuracyofthetrainedmodelonMNISTtestdatasetisabout98.2%. AlternativeTensorFlowInstallation IfyouarehavingproblemswithCUDAlibraries(ietfcouldnotloadorfindlibcudart.so.10.X),TensorFlowandCUDAlibrariescanbeinstalledtogetherusingconda: pipuninstalltensorflow-gpu condainstall-canacondatensorflow-gpu AdvancedDeepLearningwithTensorFlow2andKerascodeexamplesusedinthebook. Chapter1-Introduction MLPonMNIST CNNonMNIST RNNonMNIST Chapter2-DeepNetworks FunctionalAPIonMNIST Y-NetworkonMNIST ResNetv1andv2onCIFAR10 DenseNetonCIFAR10 Chapter3-AutoEncoders DenoisingAutoEncoders Sampleoutputsforrandomdigits: ColorizationAutoEncoder Sampleoutputsforrandomcifar10images: Chapter4-GenerativeAdversarialNetwork(GAN) DeepConvolutionalGAN(DCGAN) Radford,Alec,LukeMetz,andSoumithChintala."Unsupervisedrepresentationlearningwithdeepconvolutionalgenerativeadversarialnetworks."arXivpreprintarXiv:1511.06434(2015). Sampleoutputsforrandomdigits: Conditional(GAN) Mirza,Mehdi,andSimonOsindero."Conditionalgenerativeadversarialnets."arXivpreprintarXiv:1411.1784(2014). Sampleoutputsfordigits0to9: Chapter5-ImprovedGAN WassersteinGAN(WGAN) Arjovsky,Martin,SoumithChintala,andLéonBottou."WassersteinGAN."arXivpreprintarXiv:1701.07875(2017). Sampleoutputsforrandomdigits: LeastSquaresGAN(LSGAN) Mao,Xudong,etal."Leastsquaresgenerativeadversarialnetworks."2017IEEEInternationalConferenceonComputerVision(ICCV).IEEE,2017. Sampleoutputsforrandomdigits: AuxiliaryClassifierGAN(ACGAN) Odena,Augustus,ChristopherOlah,andJonathonShlens."ConditionalimagesynthesiswithauxiliaryclassifierGANs.Proceedingsofthe34thInternationalConferenceonMachineLearning,Sydney,Australia,PMLR70,2017." Sampleoutputsfordigits0to9: Chapter6-GANwithDisentangledLatentRepresentations InformationMaximizingGAN(InfoGAN) Chen,Xi,etal."Infogan:Interpretablerepresentationlearningbyinformationmaximizinggenerativeadversarialnets." AdvancesinNeuralInformationProcessingSystems.2016. Sampleoutputsfordigits0to9: StackedGAN Huang,Xun,etal."Stackedgenerativeadversarialnetworks."IEEEConferenceonComputerVisionandPatternRecognition(CVPR).Vol.2.2017 Sampleoutputsfordigits0to9: Chapter7-Cross-DomainGAN CycleGAN Zhu,Jun-Yan,etal."UnpairedImage-to-ImageTranslationUsingCycle-ConsistentAdversarialNetworks."2017IEEEInternationalConferenceonComputerVision(ICCV).IEEE,2017. Sampleoutputsforrandomcifar10images: SampleoutputsforMNISTtoSVHN: Chapter8-VariationalAutoencoders(VAE) VAEMLPMNIST VAECNNMNIST ConditionalVAEandBetaVAE Kingma,DiederikP.,andMaxWelling."Auto-encodingVariationalBayes."arXivpreprintarXiv:1312.6114(2013). Sohn,Kihyuk,HonglakLee,andXinchenYan."Learningstructuredoutputrepresentationusingdeepconditionalgenerativemodels."AdvancesinNeuralInformationProcessingSystems.2015. I.Higgins,L.Matthey,A.Pal,C.Burgess,X.Glorot,M.Botvinick,S.Mohamed,andA.Lerchner.β-VAE:Learningbasicvisualconceptswithaconstrainedvariationalframework.ICLR,2017. GeneratedMNISTbynavigatingthelatentspace: Chapter9-DeepReinforcementLearning Q-Learning Q-LearningonFrozenLakeEnvironment DQNandDDQNonCartpoleEnvironment Mnih,Volodymyr,etal."Human-levelcontrolthroughdeepreinforcementlearning."Nature518.7540(2015):529 DQNonCartpoleEnvironment: Chapter10-PolicyGradientMethods REINFORCE,REINFORCEwithBaseline,Actor-Critic,A2C SuttonandBarto,ReinforcementLearning:AnIntroduction Mnih,Volodymyr,etal."Asynchronousmethodsfordeepreinforcementlearning."Internationalconferenceonmachinelearning.2016. PolicyGradientonMountainCarContinuousEnvironment: Chapter11-ObjectDetection Single-ShotDetection Single-ShotDetectionon3Objects Chapter12-SemanticSegmentation FCN PSPNet SemanticSegmentation Chapter13-UnsupervisedLearningusingMutualInformation InvariantInformationClustering MINE:MutualInformationEstimation MINE Citation Ifyoufindthisworkuseful,pleasecite: @book{atienza2020advanced, title={AdvancedDeepLearningwithTensorFlow2andKeras:ApplyDL,GANs,VAEs,deepRL,unsupervisedlearning,objectdetectionandsegmentation,andmore}, author={Atienza,Rowel}, year={2020}, publisher={PacktPublishingLtd} } About AdvancedDeepLearningwithKeras,publishedbyPackt Topics reinforcement-learning deep-learning keras gan autoencoder vae Resources Readme License MITlicense Stars 1.3k stars Watchers 59 watching Forks 797 forks Releases Noreleasespublished Packages0 Nopackagespublished Contributors5 Languages Python 100.0% Youcan’tperformthatactionatthistime. Yousignedinwithanothertaborwindow.Reloadtorefreshyoursession. Yousignedoutinanothertaborwindow.Reloadtorefreshyoursession.