import os os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' import tensorflow as tf # Import dataset for experiment from tensorflow.examples.tutorials.mnist import input_data mnist = input_data.read_data_sets('MNIST_data', one_hot=True) def weight_variable(shape): # weight initialization 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') sess = tf.InteractiveSession() x = tf.placeholder(tf.float32, [None, 784]) # None means a dimension can be of any length y_ = tf.placeholder(tf.float32, [None, 10]) # First Convolutional Layer W_conv1 = weight_variable([5,5,1,32]) # heights x widths x channels x number of filters b_conv1 = bias_variable([32]) # The convolution will compute 32 features for each 5x5 patch. Its weight tensor will have a shape of [5, 5, 1, 32]. # The first two dimensions are the patch size, the next is the number of input channels, and the last is the number # of output channels. We will also have a bias vector with a component for each output channel. x_image = tf.reshape(x, [-1,28,28,1]) # segmenting the image A: and then 1 image was segmented to several images # number of images(1) x heights(28) x widths(28) x channels(1) h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1) # size of each image is [28,28] with 32 channels now h_pool1 = max_pool_2x2(h_conv1) # size of each image is [14,14] with 32 channels now # Second Convolutional Layer 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) # size of each image is [14,14] with 64 channels now h_pool2 = max_pool_2x2(h_conv2) # size of each image is [7,7] with 64 channels now # Densely Connected Layer 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) # Dropout # To reduce overfitting, we will apply dropout before the readout layer. We create a placeholder for the probability # that a neuron's output is kept during dropout. This allows us to turn dropout on during training, and turn it off # during testing. # TensorFlow's tf.nn.dropout op automatically handles scaling neuron outputs in addition to masking them, so dropout # just works without any additional scaling. keep_prob = tf.placeholder(tf.float32) h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob) # Readout Layer W_fc2 = weight_variable([1024, 10]) b_fc2 = bias_variable([10]) y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2 # Train and Evaluate the Model cross_entropy = tf.reduce_mean( tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y_conv)) train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy) correct_prediction = tf.equal(tf.argmax(y_,1),tf.argmax(y_conv,1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) sess.run(tf.global_variables_initializer()) for i in range(6000): batch = mnist.train.next_batch(50) if i%100 == 0: train_accuracy = accuracy.eval(feed_dict={x: batch[0], y_: batch[1], keep_prob: 1.0}) print ("step %d, training accuracy %g"%(i, train_accuracy)) train_step.run(feed_dict={x: batch[0], y_:batch[1], keep_prob: 0.5}) print("test accuracy %g"%accuracy.eval(feed_dict={x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0})) # ==================================== # Note here: # session: the connection between tensorflow and the C++ backend # common usage for Tensorflow: # - first create a graph # - then launch the graph in a session # weight initialization for CNN # - with a small amount of noise for symmetry breaking & to prevent 0 gradients # - when using ReLU neuron: initialize them with a slightly positive initial bias to avoid dead neurons
tensorflow 官网教程 - Deep MNIST for Experts - 代码及注解
猜你喜欢
转载自blog.csdn.net/tsinghuahui/article/details/73008519
今日推荐
周排行