TensorFlow教程04：MNIST实验——源码和运行结果

假定您已经安装好了TensorFlow，这里放了第一个MNIST实验的代码和参考结果，你可以直接运行验证。

源码

[python] view plain copy

print?

1. #!/usr/bin/python  
2. import tensorflow as tf  
3. import sys  
4. from tensorflow.examples.tutorials.mnist import input_data  
5.   
6. mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)  
7.   
8. x = tf.placeholder("float", [None, 784])  
9. W = tf.Variable(tf.zeros([784,10]))  
10. b = tf.Variable(tf.zeros([10]))  
11.   
12. y = tf.nn.softmax(tf.matmul(x,W) + b)  
13.   
14. y_ = tf.placeholder("float", [None,10])  
15. cross_entropy = -tf.reduce_sum(y_*tf.log(y))  
16.   
17. train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)  
18.   
19. init = tf.global_variables_initializer()  
20. sess = tf.Session()  
21. sess.run(init)  
22.   
23. for i in range(1000):  
24.   if i % 20 == 0:  
25.     sys.stdout.write('.')  
26.   batch_xs, batch_ys = mnist.train.next_batch(100)  
27.   sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys})  
28. print ""  
29.   
30. correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1))  
31. accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))  
32.   
33. print sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels})  

#!/usr/bin/pythonimport tensorflow as tfimport sysfrom tensorflow.examples.tutorials.mnist import input_datamnist = input_data.read_data_sets("MNIST_data/", one_hot=True)x = tf.placeholder("float", [None, 784])W = tf.Variable(tf.zeros([784,10]))b = tf.Variable(tf.zeros([10]))y = tf.nn.softmax(tf.matmul(x,W) + b)y_ = tf.placeholder("float", [None,10])cross_entropy = -tf.reduce_sum(y_*tf.log(y))train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)init = tf.initialize_all_variables()sess = tf.Session()sess.run(init)for i in range(1000):  if i % 20 == 0:    sys.stdout.write('.')  batch_xs, batch_ys = mnist.train.next_batch(100)  sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys})print ""correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1))accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))print sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels})

运行结果

[plain] view plain copy

print?

1. Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.  
2. Extracting MNIST_data/train-images-idx3-ubyte.gz  
3. Successfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.  
4. Extracting MNIST_data/train-labels-idx1-ubyte.gz  
5. Successfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.  
6. Extracting MNIST_data/t10k-images-idx3-ubyte.gz  
7. Successfully downloaded t10k-labels-idx1-ubyte.gz 4542 bytes.  
8. Extracting MNIST_data/t10k-labels-idx1-ubyte.gz  
9. ..................................................  
10. 0.9177  

Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.Extracting MNIST_data/train-images-idx3-ubyte.gzSuccessfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.Extracting MNIST_data/train-labels-idx1-ubyte.gzSuccessfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.Extracting MNIST_data/t10k-images-idx3-ubyte.gzSuccessfully downloaded t10k-labels-idx1-ubyte.gz 4542 bytes.Extracting MNIST_data/t10k-labels-idx1-ubyte.gz..................................................0.9177