TensorFlowを使ってシンプルなニューラルネットワークを作ってみる

tensorflow_neural_mnist.py

#tensorflowとインポート。MNISTデータをロード
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist= input_data.read_data_sets('MNIST_data',one_hot=True)

#プレースホルダー作成
X = tf.placeholder(tf.float32, shape=[None, 784])
Y = tf.placeholder(tf.float32, shape=[None, 10])

#重み(w)とバイアス(b)
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))

#トレーニングのためのパラメータを設定
batch_size = 100
learning_rate = 0.01
training_epochs = 10

y = tf.nn.softmax(tf.matmul(X,W) + b)

#コスト関数：クロスエントロピー誤差
cross_entropy = tf.reduce_mean(-tf.reduce_sum(Y * tf.log(y), reduction_indices=[1]))

#パラメータの精度を計算
correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(Y,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

#トレーニング（勾配降下法）
train_model = tf.train.GradientDescentOptimizer(learning_rate).minimize(cross_entropy)

with tf.Session() as sess:
    sess.run(tf.initialize_all_variables())

    for epoch in range(training_epochs) :
        batch_count = int(mnist.train.num_examples/batch_size)

        for i in range(batch_count):
            batch_x, batch_y = mnist.train.next_batch(batch_size)
            sess.run([train_model], feed_dict={X: batch_x, Y: batch_y})
            
    print "Accuracy: ", accuracy.eval(session = sess ,feed_dict={X: mnist.test.images, Y: mnist.test.labels})