def definite_model_no_GloVe():
title = "不使用预训练词嵌入"
model = Sequential()
model.add(Embedding(max_words, embedding_dim, input_length=max_len))
model.add(Flatten())
model.add(Dense(32, activation=relu))
model.add(Dense(1, activation=sigmoid))
model.summary()
history = train_model(model)
plot_classes_results(history, title, epochs)
print(title + "评估测试集", model.evaluate(x_test, y_test))
def definite_dense():
title = "将 Dense 应用于 IMDB"
model = Sequential(name=title)
model.add(Dense(16, activation=relu, input_shape=(500, )))
model.add(Dense(16, activation=relu))
model.add(Dense(1, activation=sigmoid))
model.summary()
model.compile(optimizer=rmsprop(lr=0.001),
loss=binary_crossentropy,
metrics=[binary_accuracy])
history = train_model(model, x_train, y_train)
plot_classes_results(history, title, epochs)
print(title + "评估测试集",
model.evaluate(x_test, y_test, verbose=2, use_multiprocessing=True))
pass
def definite_rnn():
title = "将 SimpleRNN 应用于 IMDB "
model = Sequential(name=title)
model.add(Embedding(max_features, 64))
model.add(SimpleRNN(64))
model.add(Dense(1, activation=sigmoid))
model.summary()
model.compile(optimizer=rmsprop(lr=0.001),
loss=binary_crossentropy,
metrics=[binary_accuracy])
history = train_model(model, x_train, y_train)
plot_classes_results(history, title, epochs)
print(title + "评估测试集",
model.evaluate(x_test, y_test, verbose=2, use_multiprocessing=True))
pass
def simple_conv1d():
print("Listing 6.46：在 IMDB 数据上训练并且评估一个简单的一维卷积神经网络")
model = Sequential(name = "简单的一维卷积神经网络")
model.add(Embedding(max_features, embedding_size, input_length = max_len))
model.add(Conv1D(32, 7, activation = relu))
model.add(MaxPooling1D(5))
model.add(Conv1D(32, 7, activation = relu))
model.add(GlobalMaxPooling1D())
model.add(Dense(1))
model.summary()
model.compile(optimizer = rmsprop(lr = 1e-4), loss = binary_crossentropy, metrics = [binary_accuracy])
history = model.fit(x_train, y_train, epochs = epochs, batch_size = batch_size,
validation_split = validation_split, verbose = verbose, use_multiprocessing = True)
title = "应用简单的一维卷积神经网络在 IMDB 数据集"
plot_classes_results(history, title, epochs)
pass
def definite_model_use_GloVe():
# Listing 6.12 模型定义
title = "使用预训练词嵌入"
model = Sequential()
model.add(Embedding(max_words, embedding_dim, input_length=max_len))
model.add(Flatten())
model.add(Dense(32, activation=relu))
model.add(Dense(1, activation=sigmoid))
# Listing 6.13 将 GloVe 训练好的词嵌入加载到 Embedding 层中
model.layers[0].set_weights([embedding_matrix])
model.layers[0].trainable = False
model.summary()
history = train_model(model)
plot_classes_results(history, title, epochs)
model.save_weights('pre_trained_glove_model.h5')
model.load_weights('pre_trained_glove_model.h5')
print(title + "评估测试集", model.evaluate(x_test, y_test))
def bidirectional_GRU():
print("Listing 6.44：训练并且评估一个双向 GRU")
title = "使用双向 GRU 模型"
model = Sequential(name=title)
model.add(Embedding(max_features, 32))
model.add(Bidirectional(GRU(32)))
model.add(Dense(1, activation=sigmoid))
model.summary()
model.compile(optimizer=rmsprop(lr=0.001),
loss=binary_crossentropy,
metrics=[binary_accuracy])
history = model.fit(x_train,
y_train,
epochs=epochs,
batch_size=batch_size,
validation_split=validation_split,
verbose=verbose,
use_multiprocessing=True)
plot_classes_results(history, title, epochs)
def reversed_lstm():
print("Listing 6.42：使用逆序序列训练并且评估一个 LSTM 模型")
title = "使用逆序序列的 LSTM 模型"
model = Sequential(name=title)
model.add(Embedding(max_features, 128))
model.add(LSTM(32))
model.add(Dense(1, activation=sigmoid))
model.summary()
model.compile(optimizer=rmsprop(lr=0.001),
loss=binary_crossentropy,
metrics=[binary_accuracy])
history = model.fit(x_train_reversed,
y_train,
epochs=epochs,
batch_size=batch_size,
validation_split=validation_split,
verbose=verbose,
use_multiprocessing=True)
plot_classes_results(history, title, epochs)
pass