搭建神经网络流程：
1.加载训练数据，并预处理（对于图像等数据，可以直接转化为矩阵，或者通过tf.convert_to_tensor()将其转换为tensor数据类型处理）；
2.构建网络层，如conv，pool，relu，lrn，fc等，在此处需要设置相应层的权重和偏置；
比较喜欢的两种定义方式如下（以定义AlexNet为例）：
import tensorflow as tf BATCH_SIZE = 200 def bias(name, shape, bias_start=0.0, trainable=True): dtype = tf.float32 var = tf.get_variable(name, shape, tf.float32, trainable=trainable, initializer=tf.constant_initializer( bias_start, dtype=dtype)) return var def weight(name, shape, stddev=0.02, trainable=True): dtype = tf.float32 var = tf.get_variable(name, shape, tf.float32, trainable=trainable, initializer=tf.random_normal_initializer( stddev=stddev, dtype=dtype)) return var def fully_connected(value, output_shape, name='fully_connected', with_w=False): value = tf.reshape(value, [BATCH_SIZE, -1]) shape = value.get_shape().as_list() with tf.variable_scope(name): weights = weight('weights', [shape[1], output_shape], 0.02) biases = bias('biases', [output_shape], 0.0) if with_w: return tf.matmul(value, weights) + biases, weights, biases else: return tf.matmul(value, weights) + biases def relu(value, name='relu'): with tf.variable_scope(name): return tf.nn.relu(value) def conv2d(value, output_dim, k_h=5, k_w=5, strides=[1, 1, 1, 1], name='conv2d'): with tf.variable_scope(name): weights = weight('weights', [k_h, k_w, value.get_shape()[-1], output_dim]) conv = tf.nn.conv2d(value, weights, strides=strides, padding='SAME') biases = bias('biases', [output_dim]) conv = tf.reshape(tf.nn.bias_add(conv, biases), conv.get_shape()) return conv def pool(value, k_size=[1, 3, 3, 1],strides=[1, 2, 2, 1], name='pool1'): with tf.variable_scope(name): pool = tf.nn.max_pool(value, ksize=k_size, strides=strides, padding='VALID') return pool def pool_avg(value, k_size=[1, 3, 3, 1],strides=[1, 2, 2, 1], name='pool1'): with tf.variable_scope(name): pool = tf.nn.avg_pool(value, ksize=k_size, strides=strides, padding='VALID') return pool def lrn(value, depth_radius=1, alpha=5e-05, beta=0.75, name='lrn1'): with tf.variable_scope(name): norm1 = tf.nn.lrn(value, depth_radius=depth_radius, bias=1.0, alpha=alpha, beta=beta) return norm1 def discriminator(image, hashing_bits, reuse=False, name='discriminator'): with tf.name_scope(name): if reuse: tf.get_variable_scope().reuse_variables() conv1 = conv2d(image, output_dim=32, name='d_conv1') relu1 = relu(pool(conv1, name='d_pool1'), name='d_relu1') conv2 = conv2d(lrn(relu1, name='d_lrn1'), output_dim=32, name='d_conv2') relu2 = relu(pool_avg(conv2, name='d_pool2'), name='d_relu2') conv3 = conv2d(lrn(relu2, name='d_lrn2'), output_dim=64, name='d_conv3') pool3 = pool_avg(relu(conv3, name='d_relu3'), name='d_pool3') relu_ip1 = relu(fully_connected(pool3, output_shape=500, name='d_ip1'), name='d_relu4') ip2 = fully_connected(relu_ip1, output_shape=hashing_bits, name='d_ip2') return ip2
和
import tensorflow as tf import numpy as np class AlexNet(object): """Implementation of the AlexNet.""" def __init__(self, x, keep_prob, num_classes, skip_layer, weights_path='DEFAULT'): # Parse input arguments into class variables self.X = x self.NUM_CLASSES = num_classes self.KEEP_PROB = keep_prob self.SKIP_LAYER = skip_layer if weights_path == 'DEFAULT': self.WEIGHTS_PATH = 'bvlc_alexnet.npy' else: self.WEIGHTS_PATH = weights_path # Call the create function to build the computational graph of AlexNet self.create() def create(self): """Create the network graph.""" # 1st Layer: Conv (w ReLu) -> Lrn -> Pool conv1 = conv(self.X, 11, 11, 96, 4, 4, padding='VALID', name='conv1') norm1 = lrn(conv1, 2, 2e-05, 0.75, name='norm1') pool1 = max_pool(norm1, 3, 3, 2, 2, padding='VALID', name='pool1') # 2nd Layer: Conv (w ReLu) -> Lrn -> Pool with 2 groups conv2 = conv(pool1, 5, 5, 256, 1, 1, groups=2, name='conv2') norm2 = lrn(conv2, 2, 2e-05, 0.75, name='norm2') pool2 = max_pool(norm2, 3, 3, 2, 2, padding='VALID', name='pool2') # 3rd Layer: Conv (w ReLu) conv3 = conv(pool2, 3, 3, 384, 1, 1, name='conv3') # 4th Layer: Conv (w ReLu) splitted into two groups conv4 = conv(conv3, 3, 3, 384, 1, 1, groups=2, name='conv4') # 5th Layer: Conv (w ReLu) -> Pool splitted into two groups conv5 = conv(conv4, 3, 3, 256, 1, 1, groups=2, name='conv5') pool5 = max_pool(conv5, 3, 3, 2, 2, padding='VALID', name='pool5') # 6th Layer: Flatten -> FC (w ReLu) -> Dropout flattened = tf.reshape(pool5, [-1, 6*6*256]) fc6 = fc(flattened, 6*6*256, 4096, name='fc6') dropout6 = dropout(fc6, self.KEEP_PROB) # 7th Layer: FC (w ReLu) -> Dropout fc7 = fc(dropout6, 4096, 4096, name='fc7') self.dropout7 = dropout(fc7, self.KEEP_PROB) #H layer:sigmoid H = get_H(self.dropout7,4096,128,name='H') # 8th Layer: FC and return unscaled activations self.fc8 = fc(self.dropout7, 4096, self.NUM_CLASSES, relu=False, name='fc8') def load_initial_weights(self, session): """Load weights from file into network.""" # Load the weights into memory weights_dict = np.load(self.WEIGHTS_PATH, encoding='bytes').item() # Loop over all layer names stored in the weights dict for op_name in weights_dict: # Check if layer should be trained from scratch if op_name not in self.SKIP_LAYER: with tf.variable_scope(op_name, reuse=True): # Assign weights/biases to their corresponding tf variable for data in weights_dict[op_name]: # Biases if len(data.shape) == 1: var = tf.get_variable('biases', trainable=False) session.run(var.assign(data)) # Weights else: var = tf.get_variable('weights', trainable=False) session.run(var.assign(data)) def conv(x, filter_height, filter_width, num_filters, stride_y, stride_x, name, padding='SAME', groups=1): """Create a convolution layer.""" # Get number of input channels input_channels = int(x.get_shape()[-1]) # Create lambda function for the convolution convolve = lambda i, k: tf.nn.conv2d(i, k, strides=[1, stride_y, stride_x, 1], padding=padding) with tf.variable_scope(name) as scope: # Create tf variables for the weights and biases of the conv layer weights = tf.get_variable('weights', shape=[filter_height, filter_width, input_channels/groups, num_filters]) biases = tf.get_variable('biases', shape=[num_filters]) if groups == 1: conv = convolve(x, weights) # In the cases of multiple groups, split inputs & weights and else: # Split input and weights and convolve them separately input_groups = tf.split(axis=3, num_or_size_splits=groups, value=x) weight_groups = tf.split(axis=3, num_or_size_splits=groups, value=weights) output_groups = [convolve(i, k) for i, k in zip(input_groups, weight_groups)] # Concat the convolved output together again conv = tf.concat(axis=3, values=output_groups) # Add biases bias = tf.reshape(tf.nn.bias_add(conv, biases), tf.shape(conv)) # Apply relu function relu = tf.nn.relu(bias, name=scope.name) return relu def fc(x, num_in, num_out, name, relu=True): """Create a fully connected layer.""" with tf.variable_scope(name) as scope: if relu: # Create tf variables for the weights and biases weights = tf.get_variable('weights', shape=[num_in, num_out], trainable=True) biases = tf.get_variable('biases', [num_out], trainable=True) # Matrix multiply weights and inputs and add bias else: weights = tf.get_variable('weights', shape=[num_in, num_out], initializer=tf.truncated_normal_initializer(stddev=0.005), trainable=True) biases = tf.get_variable('biases', shape=[num_out], initializer=tf.constant_initializer(1.0), trainable=True) act = tf.nn.xw_plus_b(x, weights, biases, name=scope.name) if relu: # Apply ReLu non linearity relu = tf.nn.relu(act) return relu else: return act def get_H(x,num_in, num_out,name): with tf.variable_scope(name) as scope: weightd=tf.get_variable('weights',shape=[num_in, num_out], initializer=tf.truncated_normal_initializer(stddev=0.005),trainable=True) biases=tf.get_variable('biases',shape=[num_out], initializer=tf.constant_initializer(1.0),trainable=True) act=tf.nn.xw_plus_b(x,weightd,biases,name=scope.name) return tf.nn.sigmoid(act) def max_pool(x, filter_height, filter_width, stride_y, stride_x, name, padding='SAME'): """Create a max pooling layer.""" return tf.nn.max_pool(x, ksize=[1, filter_height, filter_width, 1], strides=[1, stride_y, stride_x, 1], padding=padding, name=name) def lrn(x, radius, alpha, beta, name, bias=1.0): """Create a local response normalization layer.""" return tf.nn.local_response_normalization(x, depth_radius=radius, alpha=alpha, beta=beta, bias=bias, name=name) def dropout(x, keep_prob): """Create a dropout layer.""" return tf.nn.dropout(x, keep_prob)
3.定义节点，准备接收数据：
x = tf.placeholder(tf.float32, [batch_size, 227, 227, 3]) y = tf.placeholder(tf.float32, [batch_size, num_classes]) keep_prob = tf.placeholder(tf.float32)
4.定义网络层；
5.定义损失函数；
6.选择optimizer是的损失函数最小；
7.初始化所有变量，通过sess.run(optimizer)来迭代学习。
在整个过程中可以设置CPU/GPU，模型持久化和可视化等。