1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
| '''
A logistic regression learning algorithm example using TensorFlow library.
This example is using the MNIST database of handwritten digits
(http://yann.lecun.com/exdb/mnist/)
Author: Aymeric Damien
Project: https://github.com/aymericdamien/TensorFlow-Examples/
'''
from __future__ import print_function
import tensorflow as tf
# Import MNIST data
# 本例使用MNIST，数字识别数据集
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/", one_hot=True)
# Parameters
learning_rate = 0.01
training_epochs = 25
batch_size = 100
display_step = 1
# tf Graph Input
# 数据集每个图片是28*28像素，结果是0-9共10个数字
# None保留，可以是任何数，在这一例中后续作为每批测试数据数量(batch_size)输入
x = tf.placeholder(tf.float32, [None, 784]) # mnist data image of shape 28*28=784
y = tf.placeholder(tf.float32, [None, 10]) # 0-9 digits recognition => 10 classes
# Set model weights
# 创建变量
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))
# Construct model
# 回归方程 Y = x·W + b ， 可判断出Y为shape为(None, 10)的张量
# 预测函数 pred = softmax(Y) ，对Y应用归一化指数函数，使输出概率化，
# 得到的结果看作0-9数字的预测概率，取最大值则为预测结果。
pred = tf.nn.softmax(tf.matmul(x, W) + b) # Softmax
# Minimize error using cross entropy
# 使用交叉熵作代价函数
# 知y(None,10),pred(None,10), y*tf.log(pred)对应位相乘得到(None,10)的张量；
# reduce_sum对维度alix=1求和降维，效果就是sum(p*log(1/q))，得到(None)个交叉熵，
# 即每一个单独训练集的交叉熵。
# reduce_mean对(None)个交叉熵求平均值，得到这一次训练的平均交叉熵，并以此为代价进行优化。
cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred), reduction_indices=1))
# Gradient Descent
# 对代价函数cost使用梯度下降优化W,b变量
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)
# Initialize the variables (i.e. assign their default value)
init = tf.global_variables_initializer()
# Start training
with tf.Session() as sess:
# Run the initializer
sess.run(init)
# Training cycle
# 进行training_epochs次训练
for epoch in range(training_epochs):
avg_cost = 0.
total_batch = int(mnist.train.num_examples/batch_size)
# Loop over all batches
# 每次取batch_size条数据，共需total_batch次
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
# Run optimization op (backprop) and cost op (to get loss value)
# 执行训练，并记录每次训练得到的cost以计算整个训练集的平均cost
_, c = sess.run([optimizer, cost], feed_dict={x: batch_xs,
y: batch_ys})
# Compute average loss 平均cost
avg_cost += c / total_batch
# Display logs per epoch step
if (epoch+1) % display_step == 0:
print("Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(avg_cost))
print("Optimization Finished!")
# Test model
# argmax返回pred在维度alis=1上最大值的下标，pred是(None,10)，所以得到(None)个预测值（概率最高）
# correct_prediction就是(None)个bool，是否预测正确
correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
# Calculate accuracy 准确率
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
# 将MNIST的测试数据代入，计算准确率
# .eval 与 sess.run(accuracy)相同
print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
|