tensorflow入门教程.pptx

Tensorflow入门教程;主流深度学习框架;人工神经网络;Tensorflow介绍;Tensorflow安装及编程基础;Tensor（张量）;The Computational Graph（计算图）;Session（会话）;placeholders（占位符）;Variables （变量）;tf.train;tf.estimator;MNIST实例分析;mnist.train.images 是一个形状为 [55000, 784] 的张量，第一个维度数字用来索引图片，第二个维度数字用来索引每张图片中的像素点。在此张量里的每一个元素，都表示某张图片里的某个像素的强度值，值介于0和1之间。 ;one-hot vectors数字n将表示成一个只有在第n维度（从0开始）数字为1的10维向量。?比如，标签0将表示成([1,0,0,0,0,0,0,0,0,0,0])。mnist.train.labels是一个形状 [55000, 10] 的张量;数据下载地址;权重初始化 Weight Initialization;卷积和池化 Convolution and Pooling;step1，输入层 x_image = tf.reshape(x, [-1,28,28,1]) tf.reshape(tensor, shape, name=None)调整tensor形状。 step2，第一层卷积和池化 W_conv1 = weight_variable([5, 5, 1, 32]) b_conv1 = bias_variable([32]) h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1) h_pool1 = max_pool_2x2(h_conv1) step3，第二层卷积和池化 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) ; step4，全连接层 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) step5，优化层 keep_prob = tf.placeholder(tf.float32) h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob) step6，输出层 W_fc2 = weight_variable([1024, 10]) b_fc2 = bias_variable([10]) y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2 ; step7，评估和训练 cross_entropy = -tf.reduce_sum(y_*tf.log(y_conv)) 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(20000): 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={