model.add(Conv1D(filters * 2,
1,
padding='valid',
activation='relu',
strides=1))
model.add(Conv1D(10,
1,
padding='valid',
activation='relu',
strides=1))
model.add(GlobalAveragePooling1D())
model.add(Dense(1))
model.add(Activation('sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.summary()
print('Train...')
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
validation_data=(x_val, y_val))
score, acc = model.evaluate(x_test, y_test, batch_size=batch_size)
print('Test score:', score)
print('Test accuracy:', acc)
export(model, history, tokenizer, name="sentiment_140_cnn_rnn", score=score, acc=acc)
from sklearn.feature_extraction.text import CountVectorizer
import pickle
from sklearn.naive_bayes import MultinomialNB
from datasets import sentiment_140
from utils.data_utils import export
maxlen = 100
# Get data
(x_train, y_train), (x_val, y_val), (x_test,
y_test) = sentiment_140.load_data()
# No need for train and validation for bayes
x_train = np.concatenate((x_train, x_val))
y_train = np.concatenate((y_train, y_val))
vectorizer = CountVectorizer()
# Creates vocabulary for words, and transform sentences in train to a word frequency vector.
x_train = vectorizer.fit_transform(x_train)
x_test = vectorizer.transform(x_test) # transform to frequency vector.
bayes = MultinomialNB()
bayes.fit(x_train, y_train)
acc = bayes.score(x_test, y_test)
export(bayes, vocabulary=vectorizer, acc=acc)
print('Build model...')
model = Sequential()
model.add(Embedding(max_features, embedding_size, input_length=maxlen))
model.add(Dropout(0.5))
model.add(Conv1D(filters,
kernel_size,
padding='valid',
activation='relu',
strides=1))
model.add(MaxPooling1D(pool_size=pool_size))
model.add(GRU(gru_output_size))
model.add(Dense(1))
model.add(Activation('sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
print('Train...')
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
callbacks=[EarlyStopping(patience=2)],
validation_data=(x_val, y_val))
score, acc = model.evaluate(x_test, y_test, batch_size=batch_size)
print('Test score:', score)
print('Test accuracy:', acc)
export(model, history, name="imdb_cnn_gru", score=score, acc=acc)
model.add(
Conv1D(filters, kernel_size, padding='valid', activation='relu',
strides=1))
model.add(MaxPooling1D(pool_size=pool_size))
model.add(GRU(gru_output_size))
model.add(Dense(1))
model.add(Activation('sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.summary()
print('Train...')
history = model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
validation_data=(x_val, y_val))
score, acc = model.evaluate(x_test, y_test, batch_size=batch_size)
print('Test score:', score)
print('Test accuracy:', acc)
export(model,
history,
tokenizer,
name="sentiment_140_neg_gru_lstm",
score=score,
acc=acc)
