In [ ]:

    

from keras.datasets import reuters


    

In [ ]:

    

(train_data, train_labels), (test_data, test_labels) = reuters.load_data(num_words = 10000)


    

In [ ]:

    

len(train_data)


    

In [ ]:

    

len(test_data)


    

In [ ]:

    

train_data[0]


    

In [ ]:

    

word_index = reuters.get_word_index()
reverse_word_index = dict([value, key] for key, value in word_index.items())


    

In [ ]:

    

decoded_newswire = ' '.join(reverse_word_index.get(i - 3, '?') for i in train_data[0])


    

In [ ]:

    

decoded_newswire


    

In [ ]:

    

train_labels[10]


    

In [ ]:

    

import numpy as np

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


    

In [ ]:

    

x_train = vectorize_data(train_data)


    

In [ ]:

    

x_test = vectorize_data(test_data)


    

In [ ]:

    

def one_hot_encode(labels, dimension = 46):
    results = np.zeros((len(labels), dimension))
    for i, label in enumerate(labels):
        results[i, label] = 1.
    return results


    

In [ ]:

    

one_hot_train_labels = one_hot_encode(train_labels)


    

In [ ]:

    

one_hot_test_labels = one_hot_encode(test_labels)


    

In [ ]:

    

from keras import models
from keras import layers


    

In [ ]:

    

model = models.Sequential()
model.add(layers.Dense(64, activation = "relu", input_shape = (10000,)))
model.add(layers.Dense(64, activation = "relu"))
model.add(layers.Dense(46, activation = "softmax"))


    

In [ ]:

    

model.compile(loss = "categorical_crossentropy", 
             optimizer = "rmsprop", 
             metrics = ["accuracy"])


    

In [ ]:

    

x_val = x_train[:1000]
partial_x_train = x_train[1000:]


    

In [ ]:

    

y_val = one_hot_train_labels[:1000]
partial_y_train = one_hot_train_labels[1000:]


    

In [ ]:

    

history = model.fit(partial_x_train, partial_y_train, epochs = 20, batch_size = 512, validation_data = (x_val, y_val))


    

In [ ]:

    

import matplotlib.pyplot as plt


    

In [ ]:

    

values = history.history
values.keys()


    

In [ ]:

    

loss = values["loss"]
val_loss = values["val_loss"]

epochs = range(1, len(loss) + 1)

plt.plot(epochs, loss, 'bo', label = "training loss")
plt.plot(epochs, val_loss, 'b', label = "validaion loss")
plt.title("Training and validation loss")
plt.xlabel("Epochs")
plt.ylabel("Loss")
plt.legend()


    

In [ ]:

    

plt.show()


    

In [ ]:

    

plt.clf()

acc = values["acc"]
val_acc = values["val_acc"]

plt.plot(epochs, acc, 'bo', label = "training accuracy")
plt.plot(epochs, val_acc, 'b', label = "validation accuracy")
plt.xlabel("Epochs")
plt.ylabel("Accuracy")
plt.legend()
plt.title("Training and Validation accuracy")


    

In [ ]:

    

plt.show()


    

In [ ]:

    

model = models.Sequential()
model.add(layers.Dense(64, activation = "relu", input_shape = (10000,)))
model.add(layers.Dense(64, activation = "relu"))
model.add(layers.Dense(46, activation = "softmax"))


    

In [ ]:

    

model.compile(loss = "categorical_crossentropy",
             optimizer = "rmsprop",
             metrics = ["accuracy"])


    

In [ ]:

    

model.fit(partial_x_train, partial_y_train, epochs = 9, batch_size = 512, validation_data = (x_val, y_val))


    

In [ ]:

    

results = model.evaluate(x_test, one_hot_test_labels)


    

In [ ]:

    

results


    

In [ ]:

    

import copy


    

In [ ]:

    

test_labels_copy = copy.copy(test_labels)
np.random.shuffle(test_labels_copy)
hits_array = np.array(test_labels) == np.array(test_labels_copy)
float(np.sum(hits_array)) / len(test_labels)


    

In [ ]:

    

predictions = model.predict(x_test)


    

In [ ]:

    

predictions[0].shape


    

In [ ]:

    

np.sum(predictions[0])


    

In [ ]:

    

np.argmax(predictions[0])