Basics of image classification with Keras – Towards Data Science

《Basics of image classification with Keras – Towards Data Science》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、JohnOlafenwaFollowA22yearsold,selfTaught ComputerProgrammerandDeepNeuralNetworkPractitionerwithfocusonComputerVision.LivesInNigeria.Jan19·5minreadBasicsofimageclassicationwithKerasInmypreviouspost,Idelvedintosomeofthetheoreticalconceptsunderlyingarticialneuralnetworks

2、.Inthispost,Iwouldbeexplainingsomecommonoperationsthatyouwouldfrequentlyneedinkeras.First,howtosavemodelsandusethemforpredictionlater,displayingimagesfromthedatasetandloadingimagesfromoursystemandpredictingtheirclass.FireupyourIDEifyouhaven’tdonesoandreadon.SAVINGMODELS

3、Trainingmodelsisaveryslowprocess,nobodywant’stodothateverytime,fortunately,weonlyneedtotrainourmodelonce,saveitandthenwecanloaditanytimeanduseittopredictnewimages.Kerassavesmodelsinthe.h5format,soincaseyouskippedinstallingh5pyinthersttutorialIposted,pleasrunpip3install

4、h5pyWewouldalsoneedmatplotlibtovisualizeourimage,hence,runpip3installmatplotlibHereisthecodeforthersttutorialimportkerasfromkeras.datasetsimportmnistfromkeras.layersimportDensefromkeras.modelsimportSequentialfromkeras.optimizersimportSGD(train_x,train_y),(test_x,test_y

5、)=mnist.load_data()#train_x=train_x.astype('float32')/255#test_x=test_x.astype('float32')/255print(train_x.shape)print(train_y.shape)print(test_x.shape)print(test_y.shape)train_x=train_x.reshape(60000,784)test_x=test_x.reshape(10000,784)train_y=keras.utils.to_categorica

6、l(train_y,10)test_y=keras.utils.to_categorical(test_y,10)model=Sequential()model.add(Dense(units=128,activation="relu",input_shape=(784,)))model.add(Dense(units=128,activation="relu"))model.add(Dense(units=128,activation="relu"))model.add(Dense(units=10,activation="soft

7、max"))model.compile(optimizer=SGD(0.001),loss="categorical_crossentropy",metrics=["accuracy"])model.fit(train_x,train_y,batch_size=32,epochs=10,verbose=1)accuracy=model.evaluate(x=test_x,y=test_y,batch_size=32)print("Accuracy:",accuracy[1])Tosavethemodel,simplyaddbelowa

8、ftermodel.t()model.save("mnist-model.h5")INFERENCEInferencereferstotheprocessofpredictingnewimagesusingourmod