import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
#載入數據集
mnist = input_data.read_data_sets("F:\\TensorflowProject\\MNIST_data",one_hot=True)
#每個批次的大小，訓練時一次100張放入神經網絡中訓練
batch_size = 100
#計算一共有多少個批次
n_batch = mnist.train.num_examples//batch_size
#定義兩個placeholder
x = tf.placeholder(tf.float32,[None,784])
#0-9十個數字
y = tf.placeholder(tf.float32,[None,10])
#創建一個神經網絡
W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))
prediction = tf.nn.softmax(tf.matmul(x,W)+b)
#二次代價函數
#loss = tf.reduce_mean(tf.square(y-prediction))
#交叉熵代價函數
#使用交叉熵定義代價函數，可以加快模型收斂速度
#loss = tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=prediction)
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=prediction))
#使用梯度下降法
#train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss)
train_step = tf.train.AdamOptimizer(0.01).minimize(loss) #1e-2
#初始化變量
init = tf.global_variables_initializer()
#結果存放在一個布爾型列表中
correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(prediction,1))
#求准確率
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
#
with tf.Session() as sess:
sess.run(init)
for epoch in range(21):
for batch in range(n_batch):
batch_xs,batch_ys = mnist.train.next_batch(batch_size)
sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys})
#測試准確率
acc = sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels})
print("Iter: "+str(epoch)+" ,Testing Accuracy "+str(acc))
###########運行結果
Extracting F:\TensorflowProject\MNIST_data\train-images-idx3-ubyte.gz Extracting F:\TensorflowProject\MNIST_data\train-labels-idx1-ubyte.gz Extracting F:\TensorflowProject\MNIST_data\t10k-images-idx3-ubyte.gz Extracting F:\TensorflowProject\MNIST_data\t10k-labels-idx1-ubyte.gz Iter: 0 ,Testing Accuracy 0.9221 Iter: 1 ,Testing Accuracy 0.9133 Iter: 2 ,Testing Accuracy 0.9271 Iter: 3 ,Testing Accuracy 0.9262 Iter: 4 ,Testing Accuracy 0.9299 Iter: 5 ,Testing Accuracy 0.9293 Iter: 6 ,Testing Accuracy 0.9301 Iter: 7 ,Testing Accuracy 0.9299 Iter: 8 ,Testing Accuracy 0.9287 Iter: 9 ,Testing Accuracy 0.9319 Iter: 10 ,Testing Accuracy 0.9317 Iter: 11 ,Testing Accuracy 0.9315 Iter: 12 ,Testing Accuracy 0.9307 Iter: 13 ,Testing Accuracy 0.932 Iter: 14 ,Testing Accuracy 0.9314 Iter: 15 ,Testing Accuracy 0.9316 Iter: 16 ,Testing Accuracy 0.9311 Iter: 17 ,Testing Accuracy 0.9333 Iter: 18 ,Testing Accuracy 0.9318 Iter: 19 ,Testing Accuracy 0.9318 Iter: 20 ,Testing Accuracy 0.9289