如何显示mnist中的数据(tensroflow)

Posted 2020-09-10 这个是标题，但是为什么要有标题

     在使用mnist数据集的时候，一直想看看数据中原来的图片，还有卷积层、池化层中的图片，经过不断的捣鼓，最后终于显示了出来。只看数据集中的图片用如下代码就好了：

 1 import tensorflow.examples.tutorials.mnist.input_data as input_data
 2 import numpy as np
 3 import matplotlib.pyplot as plt
 4 import pylab
 5 
 6 mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)    
 7 
 8 batch_xs, batch_ys = mnist.train.next_batch(100)        
 9 for one_pic_vic in batch_xs:
10     one_pic_arr = np.reshape(one_pic_vic,(28,28))                           
11     plt.imshow(one_pic_arr)
12     pylab.show()

 

　　batch_xs的Size是(100,784)，其中100是由batch大小决定，mnist中的每张图片本来的大小是28x28的，然后数据集中存成了1x784，所以batch_xs对应100张图片。上面的代码通过reshape把图片又转成了28x28，然后就可以通过plt.imshow()显示出来:

      如果要看卷积神经网络中的卷积层、池化层，也可以用类似的方法，不过要先使用sess.run()方法来提取出来卷积层、池化层，因为图像比较多，所以就用plt.imsave()来保存到文件中。

import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data


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


def bias_variable(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_pool_2x2(x):
    return tf.nn.max_pool(x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding=\'SAME\')


mnist = input_data.read_data_sets("data/", one_hot=True)

sess = tf.InteractiveSession()

x = tf.placeholder("float", shape=[None, 784])
y_ = tf.placeholder("float", shape=[None, 10])

W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))

W_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])

x_image = tf.reshape(x, [-1, 28, 28, 1])

h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)

h_pool1 = max_pool_2x2(h_conv1)

W_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])

h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)

# Now image size is reduced to 7*7
W_fc1 = weight_variable([7 * 7 * 64, 1024])
b_fc1 = bias_variable([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)

keep_prob = tf.placeholder("float")
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)

W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])
y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)

#cross_entropy = -tf.reduce_sum(y_ * tf.log(y_conv))
cross_entropy = tf.reduce_sum(tf.pow(y_ - y_conv,2))

train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
sess.run(tf.initialize_all_variables())

for i in range(20000000):
    batch = mnist.train.next_batch(100)
    #print(batch)
    train_accuracy = accuracy.eval(feed_dict={
        x: batch[0], y_: batch[1], keep_prob: 1.0})
    if i % 20 is 0:
        print("step %d, training accuracy %f%%" % (i, train_accuracy*100))
    train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
    
    if(i % 7 == 0):                   
        pic = batch[0][0]
        pic = pic.reshape((28,28))
        plt.imsave("mnist/pic" + str(i)  + ".jpg",np.array(pic))
        
        conv1 = sess.run(h_conv1,feed_dict={x: batch[0], y_: batch[1]})
        for k in range(32):
            conv1_ = conv1[0,0:28,0:28,k]
            plt.imsave("mnist/pic" + str(i) + "-conv1-" + str(k) + ".jpg",np.array(conv1_))
            
        pool1 = sess.run(h_pool1,feed_dict={x: batch[0], y_: batch[1]})
        for k in range(32):
            pool1_ = pool1[0,0:14,0:14,k]
            plt.imsave("mnist/pic" + str(i) + "-pool1-" + str(k) + ".jpg",np.array(pool1_))
            
        conv2 = sess.run(h_conv2,feed_dict={x: batch[0], y_: batch[1]})
        for k in range(64):
            conv2_ = conv2[0,0:14,0:14,k]
            plt.imsave("mnist/pic" + str(i) + "-conv2-" + str(k)  + ".jpg",np.array(conv2_))

print("Training finished")

print("test accuracy %.3f" % accuracy.eval(feed_dict={
    x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))

　　可以在变量空间中发现，第一个卷积层的图片大小和原图一样，都是28x28，第一个池化层大小是14x14，图像缩小了一倍，第二个卷积层大小是14x14。

 

　　保存的图片如下：

 

　　通过上述代码可以显示mnist中的数据，但是有点麻烦，可以去这个网站看看(需要FQ)，这个网站可视化了cnn的训练过程，但是准确率不高：

 

　　如果没有梯子，也可以去这个网站看看cnn的训练过程：