Thispracticalguideprovidesnearly200self-containedrecipestohelpyousolvemachinelearningchallengesyoumayencounterinyourdailywork.Ifyou’recomfortablewithPythonanditslibraries,includingpandasandscikit-learn,you’llbeabletoaddressspecificproblemssuchasloadingdata,handlingtextornumericaldata,modelselection,anddimensionalityreductionandmanyothertopics.Eachrecipeincludescodethatyoucancopyandpasteintoatoydatasettoensurethatitactuallyworks.Fromthere,youcaninsert,combine,oradaptthecodetohelpconstructyourapplication.Recipesalsoincludeadiscussionthatexplainsthesolutionandprovidesmeaningfulcontext.Thiscookbooktakesyoubeyondtheoryandconceptsbyprovidingthenutsandboltsyouneedtoconstructworkingmachinelearningapplications.You’llfindrecipesfor:Vectors,matrices,andarraysHandlingnumericalandcategoricaldata,text,images,anddatesandtimesDimensionalityreductionusingfeatureextractionorfeatureselectionModelevaluationandselectionLinearandlogicalregression,treesandforests,andk-nearestneighborsSupportvectormachines(SVM),naïveBayes,clustering,andneuralnetworksSavingandloadingtrainedmodels
1