lab4MMO

import numpy as np
import tensorflow as tf
from keras import models
from keras import layers
from keras.datasets import imdb

(training_data, training_targets), (testing_data, testing_targets) = imdb.load_data(num_words=10000)
data = np.concatenate((training_data, testing_data), axis=0)
targets = np.concatenate((training_targets, testing_targets), axis=0)


def vectorize(sequences, dimension=10000):
    results = np.zeros((len(sequences), dimension))
    for i, sequence in enumerate(sequences):
        results[i, sequence] = 1
    return results

def toFixed(numObj, digits=0):
    return f"{numObj:.{digits}f}"


print(data[0])
data = vectorize(data)
targets = np.array(targets).astype("float32")
test_x = data[:10000]
test_y = targets[:10000]
train_x = data[10000:]
train_y = targets[10000:]

model = models.Sequential()

# Входной слой
model.add(layers.Dropout(0.2, input_shape=(10000,)))
model.add(layers.Dense(50, activation="relu", input_shape=(10000,)))
# Скрытый слой и слои выбивания
model.add(layers.Dropout(0.3, input_shape=(50,)))
model.add(layers.Dense(50, activation="relu"))
# Выходной слой
model.add(layers.Dropout(0.2, input_shape=(50,)))
model.add(layers.Dense(1, activation="sigmoid"))
model.summary()

model.compile(
    optimizer="adam",
    loss="binary_crossentropy",
    metrics=["accuracy"]
)
results = model.fit(
    train_x, train_y,
    epochs=10,
    batch_size=500,
    validation_split=0.2
)
score = model.evaluate(test_x, test_y)

print("Test Потери: %.2f%%" % (score[0]*100))
print("Test Точность: %.2f%%" % (score[1]*100))

predictions = model.predict(test_x)
print("Предсказание -- на самом деле")
for i in range(10):
    print('%.6f -- %f' %(predictions[i], test_y[i]))

words = tf.keras.datasets.imdb.get_word_index(path="imdb_word_index.json")
print("Индексы слов согласно IMDB (первые 20шт)")
i = 0
for key, value in words.items():
  print(key,":", value),
  i+=1
  if(i>20): break

text = "The Shawshank Redemption is written and directed by Frank Darabont. \n" \
       "It is an adaptation of the Stephen King novella Rita Hayworth and Shawshank Redemption. \n" \
       "Starring Tim Robbins and Morgan Freeman, the film portrays the story of Andy Dufresne (Robbins), \n" \
       "a banker who is sentenced to two life sentences at Shawshank State Prison for apparently murdering his wife \n" \
       "and her lover. Andy finds it tough going but finds solace in the friendship he forms with fellow inmate \n" \
       "Ellis Red Redding (Freeman). While things start to pick up when the warden finds Andy a \n" \
       "prison job more befitting his talents as a banker. However, the arrival of another inmate is going \n" \
       "to vastly change things for all of them."

print("\nВходной текст рецензии\n\n", text,"\n\n")

print("Секвенция слов из входной рецензии")
from keras.preprocessing.text import text_to_word_sequence
result = text_to_word_sequence(text)
print(result)

for i in range(len(result)):
    if(words.get(result[i]) == None or words.get(result[i]) > 10000):
        result[i] = 0
    else:
        result[i] = words.get(result[i])

print("Длина секвенции")
print(len(result))
print("Преобразованная секвенция в индексы")
print(result)

vector = vectorize(result)

newpredict = model.predict(vector)
print('Уверенность нейросети в эмоциональности рецензии - %.6f ' %(newpredict[0]))