In this notebook, we duplicate the neural network created in the TensorFlow example in keras. I found the model in keras to be conceptually cleaner.
In [1]:
from __future__ import print_function
from __future__ import division
import numpy as np
np.random.seed(1337) # for reproducibility
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation, Flatten
from keras.layers.convolutional import Convolution2D, MaxPooling2D
from keras.utils import np_utils
Set up the parameters for the model. Nothing too exciting here.
In [2]:
batch_size = 500
nb_classes = 10
nb_epoch = 1
# input image dimensions
img_rows, img_cols = 28, 28
# number of convolutional filters to use
nb_filters = 32
nb2_filters = 64
# size of pooling area for max pooling
nb_pool = 2
# convolution kernel size
nb_conv = 5
# the data, shuffled and split between tran and test sets
(X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)
X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)
X_train = X_train.astype('float32')
X_test = X_test.astype('float32')
X_train /= 255 # normalize the data
X_test /= 255 # normalize the data
# convert class vectors to binary class matrices
Y_train = np_utils.to_categorical(y_train, nb_classes)
Y_test = np_utils.to_categorical(y_test, nb_classes)
Below we build the neural network. This is the same network used in the TensorFlow example.
In [3]:
model = Sequential()
model.add(Convolution2D(nb_filters, nb_conv, nb_conv,
border_mode='valid',
input_shape=(1, img_rows, img_cols)))
model.add(Activation('relu'))
model.add(Convolution2D(nb2_filters, nb_conv, nb_conv))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(nb_pool, nb_pool)))
model.add(Dropout(0.5))
model.add(Flatten())
model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='adadelta')
Now we just run the neural network. Note that one epoch here is a run through the entire set of training data, so it takes a while.
In [4]:
model.fit(X_train, Y_train, batch_size=batch_size, nb_epoch=nb_epoch,
show_accuracy=True, verbose=1, validation_data=(X_test, Y_test))
score = model.evaluate(X_test, Y_test, show_accuracy=True, verbose=0)
print('Test score:', score[0])
print('Test accuracy:', score[1])