import input_data
import tensorflow as tf

mnist = input_data.read_data_sets('data/', one_hot=True)

#定義迴歸模型
x = tf.placeholder(tf.float32, [None, 784])
w = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))
y = tf.matmul(x, w) + b

#定義損失函式和優化器
y_ = tf.placeholder(tf.float32, [None, 10])
#用tf.nn.softmax_cross_entropy_with_logits計算預測值y和真值y_的差值，並取均值
cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=y, labels=y_))
#採用SGD作為優化器
train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)

#使用InteractiveSession()建立互動式上下文的TensorFlow會話，可以先定義會話再定義操作
sess = tf.InteractiveSession()

tf.global_variables_initializer().run()

#Train
for _ in range(1000):
    batch_xs, batch_ys = mnist.train.next_batch(100)
    sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys})

#評估訓練好的模型
#計算預測值和真實值
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
#布林轉化為浮點，取平均值，得到準確率
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
#計算在測試集上的準確率
print(sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels}))
 

結果：  

0.9154