Complete-Data-Sciece-Toolkits
Theoverallobjectiveofthistoolkitistoprovideadofferafreecollectioofdataaalysisadmachielearigthatisspecificallysuitedfordoigdatasciece.Itspurposeistogetyoustartediamatterofmiutes.YoucaruthiscollectioseitheriJupyterotebookorpythoaloe.
FeaturesMachieLearigCross-ValidatioEvaluatigClassificatioMetricsEvaluatigClusterigMetricsEvaluatigRegressioMetricsGridSearchPreprocessigEcodigCategoricalFeaturesPreprocessigBiarizatioPreprocessigImputigMissigValuesPreprocessigNormalizatioPreprocessigStadardScalerRadomizedParameterOptimizatioNumpyAddig,Removig,adSplittigArraysSortigarraysMatrixobjectStatisticsVectorMathStructuredArraysImport,Export,Slicig,IdexigDatatofromstrigPadasCompletepadasGroupbyiPadasMappigFilterigApplyigVisualizatioBarPlotsCustomizatioMatplotlibWorkigwithImageWorkigwithtextNamigCovetiosTheamigcovectiosIfollowedis:[yyyy-mm-dd-i-project-ame-library].extetioyyyy=stadsforyearmm=stadsformothdd=stadsfordayi=myiitial,forexample:SalebaOlow=solibrary=umpy,padas,sklear,matplotlibproject-ame=eachprojectameextetio=.ipyb,.py,.htmlExample:2017-25-11-so-cross-validatio-sklear.ipybCodeSamples:CrossValidatio
fromsklear.model_selectioimportcross_val_scoremodel=SVC(kerel='liear',C=1)#let'stryitusigcvscores=cross_val_score(model,X,y,cv=5)GridSearch
fromsklear.grid_searchimportGridSearchCVparams={"_eighbors":p.arage(1,5),"metric":["euclidea","cityblock"]}grid=GridSearchCV(estimator=k,param_grid=params)grid.fit(X_trai,y_trai)prit(grid.best_score)prit(grid.best_estimator_._eighbors)PreprocessigImputigMissigValues
fromsklear.preprocessigimportImputerimpute=Imputer(missig_values=0,strategy='mea',axis=0)impute.fit_trasform(X_trai)RadomizedParameterOptimizatio
fromsklear.grid_searchimportRadomizedSearchCVparams={"_eighbors":rage(1,5),"weights":["uiform","distace"]}rsearch=RadomizedSearchCV(estimator=k,param_distributios=params,cv=4,_iter=8,radom_state=5)rsearch.fit(X_trai,y_trai)prit(rsearch.best_score_)Modelfittigsupervisedadusupervisedlearig
#supervisedlearigfromsklearimporteighborsk=eighbors.KNeighborsClassifier(_eighbors=5)k.fit(X_trai,y_trai)#usupervisedlearigfromsklear.decompositioimportPCApca=PCA(_compoets=0.95)pca_model=pca.fit_trasform(X_trai)Workigwithumpyarrays
importumpyasp#appedsvaluestoedofarrp.apped(arr,values)#isertsvaluesitoarrbeforeidex2p.isert(arr,2,values)IdexigadSlicigarrays
importumpyasp#returtheelemetatidex5arr=p.array([[1,2,3,4,5,6,7]])arr[5]#retursthe2Darrayelemetoidexarr[2,5]#assigarrayelemetoidex1thevalue4arr[1]=4#assigarrayelemetoidex[1][3]thevalue10arr[1,3]=10CreatigDataFrame
importpadasaspd#specifyvaluesforeachrowsadcolumsdf=pd.DataFrame([[4,7,10],[5,8,11],[6,9,12]],idex=[1,2,3],colums=['a','b','c'])groupbypadas
importpadasaspdimportpadasaspd#returagroupbyobject,groupedbyvaluesicolumamed'cities'df.groupby(by="Cities")hadligmissigvalues
importpadasaspd#droprowswithaycolumhavigNA/ulldata.df.dropa()#replaceallNA/ulldatawithvaluedf.filla(value)Meltfuctio
importpadasaspd#mostpadasmethodsreturaDataFramesothat#thisimprovesreadabilityofcodedf=(pd.melt(df).reame(colums={'old_ame':'ew_ame','old_ame':'ew_ame'}).query('ew_ame>=200'))Saveplot
mportmatplotlib.pyplotasplt#savesplot/figuretoimageplt.savefig('pic_ame.pg')Marker,lies
importmatplotlib.pyplotasplt#add*foreverydatapoitplt.plot(x,y,marker='*')#addsdotforeverydatapoitplt.plot(x,y,marker='.')Figures,Axis
importmatplotlib.pyplotasplt#acotaierthatcotaisallplotelemetsfig=plt.figures()#Iitializessubplotfig.add_axes()#Asubplotisaaxesoagridsystem,rows-colsuma=fig.add_subplot(222)#addssubplotfig,b=plt.subplots(rows=3,cols=2)#createssubplotax=plt.subplots(2,2)Workigwithtextplot
importmatplotlib.pyplotasplt#placestextatcoordiates1/1plt.text(1,1,'Exampletext',style='italic')#aotatethepoitwithcoordiatesxywithtextax.aotate('someaotatio',xy=(10,10))#justputmathformulaplt.title(r'$delta_i=20$',fotsize=10)
评论