1.8TF的分類

TF識別手寫體識別分類

#-*- coding: utf-8 -*-
# @Time    : 2017/12/26 15:42
# @Author  : Z
# @Email   : S
# @File    : 1.9classification.py
#該程序在windows上報錯，linux上沒問題
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
#網上下載數據包，也能夠下載好指定
#http://yann.lecun.com/exdb/mnist/
mnist = input_data.read_data_sets('D:\\BigData\\Data\\MNIST_data', one_hot=True)

print(mnist.train.num_examples)
#
def add_layer(inputs,in_size,out_size,activation_function=None):
    #定義權重--隨機生成inside和outsize的矩陣
    Weights=tf.Variable(tf.random_normal([in_size,out_size]))
    #不是矩陣，而是相似列表
    biaes=tf.Variable(tf.zeros([1,out_size])+0.1)
    Wx_plus_b=tf.matmul(inputs,Weights)+biaes
    if activation_function is  None:
        outputs=Wx_plus_b
    else:
        outputs=activation_function(Wx_plus_b)
    return outputs
def compute_accuracy(v_xs,v_ys):
    global prediction
    y_pre=sess.run(prediction,feed_dict={xs:v_xs})
    correct_prediction=tf.equal(tf.argmax(y_pre,1),tf.argmax(v_ys,1))
    accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
    result=sess.run(accuracy,feed_dict={xs:v_xs,ys:v_ys})
    return result
#添加placeholder對於輸入網絡層
xs=tf.placeholder(tf.float32,[None,784]) #28*28
ys=tf.placeholder(tf.float32,[None,10])
#增長輸出層
prediction=add_layer(xs,784,10,activation_function=tf.nn.softmax)
#定義loss損失---信息熵
cross_entropy=tf.reduce_mean(-tf.reduce_sum(ys*tf.log(prediction),reduce_indices=[1]))
train_step=tf.train.GradientDescentOptimizer(0.1).minimize(cross_entropy)


sess=tf.Session()
#變量的初始化
sess.run(tf.global_variables_initializer())

for i in range(1000):
    batch_xs,batch_ys=mnist.train.next_batch(100) #取一部分數據
    sess.run(train_step,feed_dict={xs:batch_xs,ys:batch_ys})
    if i%50:
        print (compute_accuracy(mnist.test.images,mnist.test.labels))

顯示結果