Exercicio Deep learning, Exercícios de Processo Estocástico

Baixe Exercicio Deep learning e outras Exercícios em PDF para Processo Estocástico, somente na Docsity!

25/11/2019 P2-Q 4.ipynb - Colaboratory from keras.models import Sequential from keras import layers from keras.optimizers import RMSprop max features = 10000 maxlen = 500 batch size = 32 print('Loading data...') model = Sequential() model.add(layers.GRU(32, input shape=(None, float data.shape[-1]))) model.add(layers.Dense(1)) model.compile(optimizer=RMSprop(), loss='mae') history = model.fit generator(train gen, steps per epoch=500, epochs=20, validation data=val gen, validation steps=val steps) model = Sequential() model.add(layers.Embedding(max features, 128, input length=max len)) model .add(layers.ConviD(32, 7, activation='relu')) model. add(layers.MaxPoolingiD(5)) model.add(layers.ConviD(32, 7, activation='relu')) model.add(layers.GlobalMaxPooling1D()) model.add(layers.Dense(1)) model. summary () model.compile(optimizer=RMSprop(Ir=1e-4), loss='binary crossentropy', metrics=['acc']) history = model.fit(x train, y train, epochs=10, batch size=128, validation split=0.2) plt.title('Training and validation accuracy") plt.legend() plt.figure() plt.plot(epochs, loss, 'bo', label='Training loss') plt.plot(epochs, val loss, 'b', label='Validation loss') plt.title('Training and validation loss') plt.legend() plt.show() D https://colab.research.google.com/drive/lidFdAeRJAy2atpCXD18Y. rJH3USicodTifscrollTo="V-ZKq3S106c&printMode=true