TensorFlow入门1-minist

                         y = softmax(Wx+b)

import tensorflow.examples.tutorials.mnist.input_data
from tensorflow.contrib.learn.python.learn.datasets.mnist import read_data_sets
mnist = read_data_sets('MNIST_data/', one_hot=True)

import tensorflow as tf

# 这里的None表示此张量的第一个维度可以是任何长度的
# x不是一个特定的值，而是一个占位符placeholder
x = tf.placeholder(tf.float32, [None, 784])

# W参数矩阵784行，10列
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))

# x 是个noneX784矩阵， y是个noneX10
# 这里为什么要用xW，而不是Wx,因为矩阵+b向量运算，会将b向量每个元素加到xW每一列上
# softmax 按照行来计算，一行算出来正好是对应y
y = tf.nn.softmax(tf.matmul(x,W)+b)
y_ = tf.placeholder('float', [None, 10])

#计算交叉熵
cross_entroy = -tf.reduce_sum(y_*tf.log(y))

train_step = tf.train.GradientDescentOptimizer(
    0.01).minimize(cross_entroy)

init = tf.global_variables_initializer()

sess = tf.Session()
sess.run(init)

for i in range(1000):
    batch_xs, batch_ys = mnist.train.next_batch(200)
    sess.run(train_step, feed_dict = {x:batch_xs, y_:batch_ys})

testArgmax=tf.argmax([[12,34,3],[78,21,45]],0)
init = tf.global_variables_initializer()
sess.run(init)
sess.run(testArgmax)

array([1, 0, 1])

testArgmax=tf.argmax([[12,34,3],[78,21,45]],1)
init = tf.global_variables_initializer()
sess.run(init)
sess.run(testArgmax)

array([1, 0])

correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1))

accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))

testArgmax1=tf.argmax([[12,34,3],[32,40,45]],0)
testArgmax2=tf.argmax([[12,34,3],[78,21,45]],0)
accurate = tf.equal(testArgmax1, testArgmax2)
testcast = tf.cast(accurate, 'float')
accuracy = tf.reduce_mean(testcast)
init = tf.global_variables_initializer()
sess.run(init)
sess.run([testcast,accuracy])

[array([1., 0., 1.], dtype=float32), 0.6666667]