def predict(): if PATH.get() is not None: input = dh.make_read_for_input(PATH.get()) prediction = dh.FromOneHot(model.predict(input))[0] CATIGORY.set('prediction is : ' + consts.CAT[prediction])
model.add(Conv2D(kernel_size = (3,3),filters = 32,activation = 'tanh',input_shape = consts.input_shape)) model.add(MaxPooling2D(pool_size=(3,3))) model.add(Conv2D(kernel_size = (3,3),filters = 64,activation = 'tanh')) model.add(MaxPooling2D(pool_size=(3,3))) model.add(Dropout(0.2)) model.add(Flatten()) model.add(Dense(100, activation = 'tanh')) model.add(Dense(64, activation = 'tanh')) model.add(Dense(16, activation = 'tanh')) model.add(Dense(consts.classes,activation = consts.activation)) optimizer = Adam(lr = consts.lr)#SGD(lr = consts.lr,decay=1e-6, momentum=0.9, nesterov=True) model.compile(optimizer,loss = consts.loss, metrics=["accuracy"]) ## training the model model.fit(ximages/255,xlabels,epochs=consts.epochs,shuffle=True,validation_data=(yimages,ylabels)) prediction = dh.FromOneHot(model.predict(yimages)) correct = dh.FromOneHot(ylabels) acc = (np.sum(prediction == correct)/len(list(yimages))) * 100 print('Acc :' + str(acc) + '%') model.save('emotions' + str(round(acc)) + '.h5')
model.add(Conv2D(kernel_size=(3, 3), filters=64, activation='tanh')) model.add(MaxPooling2D(pool_size=(3, 3))) model.add(Dropout(0.2)) model.add(Flatten()) model.add(Dense(100, activation='tanh')) model.add(Dense(64, activation='tanh')) model.add(Dense(16, activation='tanh')) model.add(Dense(consts.classes, activation=consts.activation)) optimizer = SGD(lr=consts.lr, decay=1e-6, momentum=0.9, nesterov=True) model.compile(optimizer, loss=consts.loss) ## training the model model.fit(x_train, y_train, epochs=consts.epochs, shuffle=True, validation_data=(x_test, y_test)) prediction = dh.FromOneHot(model.predict(x_test)) correct = dh.FromOneHot(y_test) acc = (np.sum(prediction == correct) / len(list(y_test))) * 100 print('Acc :' + str(acc) + '%') model.save('flowers' + str(round(acc)) + '.h5')