AI_Niewiarowska_lab8_zad2

import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Flatten, Conv2D, MaxPooling2D
from keras.utils import np_utils

import matplotlib.pyplot as plt

(X_train, y_train), (X_test, y_test) = mnist.load_data()

print(keras.backend.image_data_format())

X_train = X_train.reshape((X_train.shape[0], 28, 28, 1)).astype('float32')
X_test = X_test.reshape((X_test.shape[0], 28, 28, 1)).astype('float32')

X_train = X_train / 255
X_test = X_test / 255

y_train = np_utils.to_categorical(y_train)
y_test = np_utils.to_categorical(y_test)
num_classes = y_test.shape[1]

net = Sequential()

net.add(Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)))
net.add(Conv2D(64, (3, 3), activation='relu'))
net.add(MaxPooling2D(pool_size=(2, 2)))
net.add(Flatten())
net.add(Dense(128, activation='sigmoid'))
net.add(Dense(10, activation='softmax'))

net.compile(loss='categorical_crossentropy',
            optimizer=keras.optimizers.adadelta(), metrics=['accuracy'])

history = net.fit(X_train, y_train, validation_data=(
    X_test, y_test), epochs=2, verbose=2, batch_size=100)

num_images = 20
images_and_predictions = list(
    zip(X_test[:num_images], net.predict_classes(X_test[:num_images])))

for index, (image, prediction) in enumerate(images_and_predictions):
    plt.subplot(1, 20, index + 1)
    plt.axis('off')
    plt.imshow(image.reshape(28, 28), cmap=plt.get_cmap(
        'gray'), interpolation='nearest')
    plt.title(prediction)
plt.show()