Tensorflow 搭建自己的神经网络(三)

CNN实现

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
"""
Created on Mon Apr  8 02:46:09 2019

@author: xiexj
"""

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data 

mnist=input_data.read_data_sets('MNIST_data', one_hot=True)

def compute_accuracy(v_xs, v_ys):
#    global prediction
    y_pre = sess.run(prediction, feed_dict={xs:v_xs,keep_prob:1})
    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,keep_prob:1})
    return result

def weight_variable(shape):
    initial = tf.truncated_normal(shape, stddev=0.1)
    return tf.Variable(initial)

def bias_vatiable(shape):
    initial = tf.constant(0.1, shape=shape)
    return tf.Variable(initial)

def conv2d(x, W):
    return tf.nn.conv2d(x, W, strides=[1,1,1,1], padding='SAME')

def max_pooling_2x2(x):
    return tf.nn.max_pool(x, ksize=[1,2,2,1], strides=[1,2,2,1], padding='SAME')
    
# define placeholder for inputs to network
xs = tf.placeholder(tf.float32, [None, 784])
ys = tf.placeholder(tf.float32, [None, 10])
keep_prob = tf.placeholder(tf.float32)
x_image = tf.reshape(xs, [-1, 28, 28, 1])# [n_samples, 28,28,1]

## conv1 layer ##
W_conv1 = weight_variable([5,5,1,32])
b_conv1 = bias_vatiable([32])
h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)
h_pool1 = max_pooling_2x2(h_conv1)

## conv2 layer ##
W_conv2 = weight_variable([5,5,32,64])
b_conv2 = bias_vatiable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pooling_2x2(h_conv2)

## fc1 layer ##
W_fc1 = weight_variable([7*7*64, 1024])
b_fc1 = bias_vatiable([1024])
h_pool2_flat = tf.reshape(h_pool2, [-1,7*7*64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)

## fc2 layer ##
W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_vatiable([10])
prediction = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)

# the error between prediction and real data
cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction),
                                              reduction_indices=[1]))
train_step = tf.train.AdamOptimizer(0.0001).minimize(cross_entropy)
init = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init)
    for i in range(1000):
        batch_xs, batch_ys = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={xs:batch_xs,ys:batch_ys,keep_prob:0.5})
        if i%50 == 0:
            print(compute_accuracy(mnist.test.images[:1000], mnist.test.labels[:1000]))

Saver 保存读取

Tensorflow目前只能保存Varibales，而不能保存框架，所以需要重新定义一下框架，再把Varibales放进来重新学习。

import tensorflow as tf
import numpy as np

W = tf.Variable([[1,2,3],[3,4,5]], dtype = tf.float32, name = 'weight')
b = tf.Variable([[1,2,3]], dtype = tf.float32, name = 'biases')

init = tf.global_variables_initializer()

saver = tf.train.Saver()

with tf.Session() as sess:
    sess.run(init)
    save_path = saver.save(sess, "my_net/save_net.ckpt")
    print("Save to path:",save_path)

import tensorflow as tf
import numpy as np

# restore variables
tf.reset_default_graph()

W = tf.Variable(np.arange(6).reshape((2,3)), dtype=tf.float32, name='weight')
b = tf.Variable(np.arange(3).reshape((1,3)), dtype=tf.float32, name='biases')

saver = tf.train.Saver()

with tf.Session() as sess:
    saver.restore(sess, "my_net/save_net.ckpt")
    print("weight:", sess.run(W))
    print("biases:", sess.run(b))

根据教程编码会出现以下错误：NotFoundError: Tensor name “weight_1” not found in checkpoint files mynet/save_net.ckpt 

添加一行： tf.reset_default_graph() ： # 清除默认图的堆栈，并设置全局图为默认图

原因：
真正的原因是，我写的代码 保存和加载 在前后进行，在前后两次定义了

W = tf.Variable(xxx,name="weight")

相当于 在TensorFlow 图的堆栈创建了两次 name = “weight” 的变量，第二个（第n个）的实际 name 会变成 “weight_1” (“weight_n-1”)，之后我们在保存 checkpoint 中实际搜索的是 “weight_n-1” 这个变量 而不是 “weight” ，因此就会出错。

正常场景下，不会保存模型之后，马上加载（或在同一程序中加载），就不会出现这个情况，或者保存完之后 restart kernel (Spyder 中)，再进行参数加载。

参考博客：Tensorflow Saver & restore 以及报错问题 NotFoundError: "x_x" not found in checkpoint