imdb_sentiment.py

import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

import tensorflow as tf
from tensorflow.keras import datasets, layers, models, preprocessing
import tensorflow_datasets as tfds

max_len = 200
n_words = 10000
dim_embedding = 256
EPOCHS = 20
BATCH_SIZE = 500

# data handling
def load_data():
    # data load
    (X_train, y_train), (X_test, y_test) = datasets.imdb.load_data(num_words=n_words)
    # max_len padding
    X_train = preprocessing.sequence.pad_sequences(X_train,maxlen=max_len)
    X_test = preprocessing.sequence.pad_sequences(X_test,maxlen=max_len)
    return (X_train, y_train), (X_test, y_test)

# modeling
def build_model():
    model = models.Sequential()

    model.add(layers.Embedding(n_words, dim_embedding, input_length=max_len))

    model.add(layers.Dropout(0.3))

    # adaption max values at Feature vector from each n_words of features
    model.add(layers.GlobalMaxPooling1D())
    model.add(layers.Dense(128, activation='relu'))
    model.add(layers.Dropout(0.5))
    model.add(layers.Dense(1, activation='sigmoid'))
    return model

# training
(X_train, y_train), (X_test, y_test) = load_data()
model = build_model()
model.summary()

model.compile(
    optimizer = 'adam',
    loss = 'binary_crossentropy' ,
    metrics = ['accuracy']
)

score = model.fit(X_train,
                  y_train,
                  epochs = EPOCHS,
                  batch_size= BATCH_SIZE,
                  validation_data = (X_test, y_test)
                  )

score = model.evaluate(X_test, y_test, batch_size=BATCH_SIZE)
print('\nTest score:', score[0])
print('Test accuracy:', score[1])