import skimage.io  # bug. need to import this before tensorflow
import skimage.transform  # bug. need to import this before tensorflow
import tensorflow as tf
from tensorflow.python.ops import control_flow_ops
from tensorflow.python.training import moving_averagesfrom config import Configimport datetime
import numpy as np
import os
import timeMOVING_AVERAGE_DECAY = 0.9997
BN_DECAY = MOVING_AVERAGE_DECAY
BN_EPSILON = 0.001
CONV_WEIGHT_DECAY = 0.00004
CONV_WEIGHT_STDDEV = 0.1
FC_WEIGHT_DECAY = 0.00004
FC_WEIGHT_STDDEV = 0.01
RESNET_VARIABLES = 'resnet_variables'
UPDATE_OPS_COLLECTION = 'resnet_update_ops'  # must be grouped with training op
IMAGENET_MEAN_BGR = [103.062623801, 115.902882574, 123.151630838, ]tf.app.flags.DEFINE_integer('input_size', 224, "input image size")activation = tf.nn.reludef inference(x, is_training,num_classes=1000,num_blocks=[3, 4, 6, 3],  # defaults to 50-layer networkuse_bias=False, # defaults to using batch normbottleneck=True):# 定义网络中各参数c = Config()c['bottleneck'] = bottleneckc['is_training'] = tf.convert_to_tensor(is_training,dtype='bool',name='is_training')c['ksize'] = 3c['stride'] = 1c['use_bias'] = use_biasc['fc_units_out'] = num_classesc['num_blocks'] = num_blocksc['stack_stride'] = 2with tf.variable_scope('scale1'):c['conv_filters_out'] = 64c['ksize'] = 7c['stride'] = 2x = conv(x, c)x = bn(x, c)x = activation(x)with tf.variable_scope('scale2'):x = _max_pool(x, ksize=3, stride=2)c['num_blocks'] = num_blocks[0]c['stack_stride'] = 1c['block_filters_internal'] = 64x = stack(x, c)with tf.variable_scope('scale3'):c['num_blocks'] = num_blocks[1]c['block_filters_internal'] = 128assert c['stack_stride'] == 2x = stack(x, c)with tf.variable_scope('scale4'):c['num_blocks'] = num_blocks[2]c['block_filters_internal'] = 256x = stack(x, c)with tf.variable_scope('scale5'):c['num_blocks'] = num_blocks[3]c['block_filters_internal'] = 512x = stack(x, c)# post-netx = tf.reduce_mean(x, reduction_indices=[1, 2], name="avg_pool")if num_classes != None:with tf.variable_scope('fc'):x = fc(x, c)return x# 图像转换，从RGB转到BGR
def _imagenet_preprocess(rgb):"""Changes RGB [0,1] valued image to BGR [0,255] with mean subtracted."""red, green, blue = tf.split(3, 3, rgb * 255.0)bgr = tf.concat(3, [blue, green, red])bgr -= IMAGENET_MEAN_BGRreturn bgr# 损失函数
def loss(logits, labels):cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(logits, labels)cross_entropy_mean = tf.reduce_mean(cross_entropy)regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)loss_ = tf.add_n([cross_entropy_mean] + regularization_losses)tf.scalar_summary('loss', loss_)return loss_def stack(x, c):for n in range(c['num_blocks']):s = c['stack_stride'] if n == 0 else 1c['block_stride'] = swith tf.variable_scope('block%d' % (n + 1)):x = block(x, c)return x# 定义网络模块
def block(x, c):filters_in = x.get_shape()[-1]# Note: filters_out isn't how many filters are outputed. # That is the case when bottleneck=False but when bottleneck is # True, filters_internal*4 filters are outputted. filters_internal is how many filters# the 3x3 convs output internally.m = 4 if c['bottleneck'] else 1filters_out = m * c['block_filters_internal']shortcut = x  # branch 1c['conv_filters_out'] = c['block_filters_internal']if c['bottleneck']:with tf.variable_scope('a'):c['ksize'] = 1c['stride'] = c['block_stride']x = conv(x, c)x = bn(x, c)x = activation(x)with tf.variable_scope('b'):x = conv(x, c)x = bn(x, c)x = activation(x)with tf.variable_scope('c'):c['conv_filters_out'] = filters_outc['ksize'] = 1assert c['stride'] == 1x = conv(x, c)x = bn(x, c)else:with tf.variable_scope('A'):c['stride'] = c['block_stride']assert c['ksize'] == 3x = conv(x, c)x = bn(x, c)x = activation(x)with tf.variable_scope('B'):c['conv_filters_out'] = filters_outassert c['ksize'] == 3assert c['stride'] == 1x = conv(x, c)x = bn(x, c)with tf.variable_scope('shortcut'):if filters_out != filters_in or c['block_stride'] != 1:c['ksize'] = 1c['stride'] = c['block_stride']c['conv_filters_out'] = filters_outshortcut = conv(shortcut, c)shortcut = bn(shortcut, c)return activation(x + shortcut)# BN:批规范化
def bn(x, c):x_shape = x.get_shape()params_shape = x_shape[-1:]if c['use_bias']:bias = _get_variable('bias', params_shape,initializer=tf.zeros_initializer)return x + biasaxis = list(range(len(x_shape) - 1))beta = _get_variable('beta',params_shape,initializer=tf.zeros_initializer)gamma = _get_variable('gamma',params_shape,initializer=tf.ones_initializer)moving_mean = _get_variable('moving_mean',params_shape,initializer=tf.zeros_initializer,trainable=False)moving_variance = _get_variable('moving_variance',params_shape,initializer=tf.ones_initializer,trainable=False)# These ops will only be preformed when training.mean, variance = tf.nn.moments(x, axis)update_moving_mean = moving_averages.assign_moving_average(moving_mean,mean, BN_DECAY)update_moving_variance = moving_averages.assign_moving_average(moving_variance, variance, BN_DECAY)tf.add_to_collection(UPDATE_OPS_COLLECTION, update_moving_mean)tf.add_to_collection(UPDATE_OPS_COLLECTION, update_moving_variance)mean, variance = control_flow_ops.cond(c['is_training'], lambda: (mean, variance),lambda: (moving_mean, moving_variance))x = tf.nn.batch_normalization(x, mean, variance, beta, gamma, BN_EPSILON)#x.set_shape(inputs.get_shape()) ??return x# 定义全连接层
def fc(x, c):num_units_in = x.get_shape()[1]num_units_out = c['fc_units_out']weights_initializer = tf.truncated_normal_initializer(stddev=FC_WEIGHT_STDDEV)weights = _get_variable('weights',shape=[num_units_in, num_units_out],initializer=weights_initializer,weight_decay=FC_WEIGHT_STDDEV)biases = _get_variable('biases',shape=[num_units_out],initializer=tf.zeros_initializer)x = tf.nn.xw_plus_b(x, weights, biases)return xdef _get_variable(name,shape,initializer,weight_decay=0.0,dtype='float',trainable=True):"A little wrapper around tf.get_variable to do weight decay and add to""resnet collection"if weight_decay > 0:regularizer = tf.contrib.layers.l2_regularizer(weight_decay)else:regularizer = Nonecollections = [tf.GraphKeys.VARIABLES, RESNET_VARIABLES]return tf.get_variable(name,shape=shape,initializer=initializer,dtype=dtype,regularizer=regularizer,collections=collections,trainable=trainable)# 定义卷积层
def conv(x, c):ksize = c['ksize']stride = c['stride']filters_out = c['conv_filters_out']filters_in = x.get_shape()[-1]shape = [ksize, ksize, filters_in, filters_out]initializer = tf.truncated_normal_initializer(stddev=CONV_WEIGHT_STDDEV)weights = _get_variable('weights',shape=shape,dtype='float',initializer=initializer,weight_decay=CONV_WEIGHT_DECAY)return tf.nn.conv2d(x, weights, [1, stride, stride, 1], padding='SAME')# 定义池化层
def _max_pool(x, ksize=3, stride=2):return tf.nn.max_pool(x,ksize=[1, ksize, ksize, 1],strides=[1, stride, stride, 1],padding='SAME')