Tensorflow学习之卷积神经网络实现（六）

  本次主要在Tensorflow中实现ResNetV2，通常认为神经网络的深度对其性能非常重要，但是网络越深其训练那度越大，于Resnet相似的Highway Network的目标就是解决极深的神经网络难以训练的问题。修改了每一层的激活函数，此前的激活函数只是对输入做一个非线性变换y=H(x,Wh),而Highway Network则允许保留一定比例的原始输入x，即y=H(x,Wh)T(x,Wt)+x*C(x,Wc)，其中T为变换系数，C为保留系数，论文中令C=1-T，这样前面一层的信息，有一定的比例可以不经过矩阵乘法和非线性变换，直接传输到下一层，主要通过gating units学习如何控制网络中的信息流，即学习原始信息应保留的比例。
  而Resnet最初的灵感出自：在不断加深神经网络的深度时，会出现一个Degradation问题，即准确率会先上升然后达到饱和，在持续增加深度会导致准确率下降，这并不是过拟合的问题，因为训练和验证误差都增大。假设输入为x，期望输出为H（x），那我们需要学习的就是F(x)=H(x)-x，也就是残差。通过直接将输入信息绕道传到输出，保护信息的完整性，整个网络则只需要学习输入、输出差别的那一部分，简化学习目标和难度。而ResnetV2则是发现前馈和反馈信号可以直接传输，因此将非线性激活函数替换为Identity Mapping（y=x），同时在每一层都适用了Batch Normalization，这样处理之后，新的残差学习单元将比以前更容易训练且泛化性更强。

import collectionsimport tensorflow as tfslim = tf.contrib.slimfrom datetime import datetimeimport timeimport math#使用clooection.namedtuple设计resnet的Block模块组并创建类#args中三个参数代表(depth,depth_bottleneck,stride)class Block(collections.namedtuple('Bolck',['scope','unit_fn','args'])):    'A named tuple tuple describing a Resnet block'def subsample(inputs,factor,scope=None):    if factor == 1:        return inputs    else:        return slim.max_pool2d(inputs,[1,1],stride=factor,scope=scope)def conv2d_same(inputs,num_outputs,kernel_size,stride,scope = None):    if stride == 1:        return slim.conv2d(inputs,num_outputs,kernel_size,stride=1,padding = 'SAME',scope=scope)    else:        pad_total = kernel_size - 1        pad_beg = pad_total//2        pad_end = pad_total-pad_beg        inputs = tf.pad(inputs,[[0,0],[pad_beg,pad_end],[pad_beg,pad_end],[0,0]])        return slim.conv2d(inputs,num_outputs,kernel_size,stride=stride,padding = 'VALID',scope = scope)#堆叠Block函数@slim.add_arg_scopedef stack_blocks_dense(net,blocks,outputs_collections=None):    for block in blocks:        with tf.variable_scope(block.scope,'block',[net]) as sc:            for i ,unit in enumerate(block.args):                with tf.variable_scope('unit_%d'%(i+1),values=[net]):                    unit_depth,unit_depth_bottleneck,unit_stride = unit                    net = block.unit_fn(net,depth=unit_depth,depth_bottleneck=unit_depth_bottleneck,stride=unit_stride)                net = slim.utils.collect_named_outputs(outputs_collections,sc.name,net)    return net#Resnet的arg_scopedef resnet_arg_scope(is_training=True,weight_decay = 0.0001,batch_norm_decay=0.997,batch_norm_epsilon = 1e-5,batch_norm_scale = True):    batch_norm_params = {'is_training':is_training,'decay':batch_norm_decay,'epsilon':batch_norm_epsilon,'scale':batch_norm_scale,'updates_collections':tf.GraphKeys.UPDATE_OPS,}    with slim.arg_scope([slim.conv2d],                        weights_regularizer=slim.l2_regularizer(weight_decay),                        weights_initializer=slim.variance_scaling_initializer(),                        activation_fn=tf.nn.relu,                        normalizer_fn=slim.batch_norm,                        normalizer_params=batch_norm_params):        with slim.arg_scope([slim.batch_norm],**batch_norm_params):            with slim.arg_scope([slim.max_pool2d],padding='SAME') as arg_sc:                return arg_sc#核心的bottleneck残差单元@slim.add_arg_scopedef bottleneck(inputs,depth,depth_bottleneck,stride,outputs_collections=None,scope=None):    with tf.variable_scope(scope,'bottleneck_v2',[inputs]) as sc:        depth_in = slim.utils.last_dimension(inputs.get_shape(),min_rank=4)        preact = slim.batch_norm(inputs,activation_fn=tf.nn.relu,scope='preact')        if depth==depth_in:            shortcut=subsample(inputs,stride,'shortcut')        else:            shortcut = slim.conv2d(preact,depth,[1,1],stride=stride,normalizer_fn=None,activation_fn=None,scope='shortcut')        residual = slim.conv2d(preact,depth_bottleneck,[1,1],stride=1,scope='conv1')        residual = conv2d_same(residual,depth_bottleneck,3,stride,scope='conv2')        residual = slim.conv2d(residual,depth,[1,1],stride=1,normalizer_fn=None,activation_fn=None,scope='conv3')        output=shortcut+residual        return slim.utils.collect_named_outputs(outputs_collections,sc.name,output)#主函数def resnet_v2(inputs,blocks,num_classes=None,global_pool=True,include_root_block=True,reuse=None,scope=None):    with tf.variable_scope(scope,'resnet_v2',[inputs],reuse=reuse) as sc:        end_points_collection = sc.original_name_scope+'_end_points'        with slim.arg_scope([slim.conv2d,bottleneck,stack_blocks_dense],outputs_collections=end_points_collection):            net =inputs            if include_root_block:                with slim.arg_scope([slim.conv2d],activation_fn=None,normalizer_fn=None):                    net = conv2d_same(net,64,7,stride=2,scope='conv1')                net=slim.max_pool2d(net,[3,3],stride=2,scope='pool1')            net=stack_blocks_dense(net,blocks)            net=slim.batch_norm(net,activation_fn=tf.nn.relu,scope='postnorm')            if global_pool:                net = tf.reduce_mean(net,[1,2],name='pool5',keep_dims=True)            if num_classes is not None:                net = slim.conv2d(net,num_classes,[1,1],activation_fn=None,normalizer_fn=None,scope='logits')            end_points = slim.utils.convert_collection_to_dict(end_points_collection)            if num_classes is not None:                end_points['predictions'] = slim.softmax(net,scope='predictions')            return net,end_points#设计曾数分别为50，101，152，200的resnetdef resnet_v2_50(inputs,num_classes=None,global_pool=True,reuse=None,scope='resnet_v2_50'):    blocks=[        Block('block1',bottleneck,[(256,64,1)]*2+[(256,64,2)]),        Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]),        Block('block3', bottleneck, [(1024, 256, 1)] * 5 + [(1024, 256, 2)]),        Block('block4', bottleneck, [(2048, 512, 1)] * 3 )]    return resnet_v2(inputs,blocks,num_classes,global_pool,include_root_block=True,reuse=reuse,scope=scope)def resnet_v2_101(inputs,num_classes=None,global_pool=True,reuse=None,scope='resnet_v2_101'):    blocks=[        Block('block1',bottleneck,[(256,64,1)]*2+[(256,64,2)]),        Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]),        Block('block3', bottleneck, [(1024, 256, 1)] * 22 + [(1024, 256, 2)]),        Block('block4', bottleneck, [(2048, 512, 1)] * 3 )]    return resnet_v2(inputs,blocks,num_classes,global_pool,include_root_block=True,reuse=reuse,scope=scope)def resnet_v2_152(inputs,num_classes=None,global_pool=True,reuse=None,scope='resnet_v2_152'):    blocks=[        Block('block1',bottleneck,[(256,64,1)]*2+[(256,64,2)]),        Block('block2', bottleneck, [(512, 128, 1)] * 7 + [(512, 128, 2)]),        Block('block3', bottleneck, [(1024, 256, 1)] * 35 + [(1024, 256, 2)]),        Block('block4', bottleneck, [(2048, 512, 1)] * 3 )]    return resnet_v2(inputs,blocks,num_classes,global_pool,include_root_block=True,reuse=reuse,scope=scope)def resnet_v2_200(inputs,num_classes=None,global_pool=True,reuse=None,scope='resnet_v2_200'):    blocks=[        Block('block1',bottleneck,[(256,64,1)]*2+[(256,64,2)]),        Block('block2', bottleneck, [(512, 128, 1)] * 23 + [(512, 128, 2)]),        Block('block3', bottleneck, [(1024, 256, 1)] * 35 + [(1024, 256, 2)]),        Block('block4', bottleneck, [(2048, 512, 1)] * 3 )]    return resnet_v2(inputs,blocks,num_classes,global_pool,include_root_block=True,reuse=reuse,scope=scope)def time_tensorflow_run(session,target,info_string):    num_steps_burn_in = 10 #预热轮数：给程序热身，头几轮迭代有显存加载，cache命中等问题因此可以跳过，只考量10论迭代之后计算时间    total_duration = 0.0 #总时间    total_duration_squared = 0.0#平方和用以计算方差    for i in range(num_batches + num_steps_burn_in):        start_time = time.time() #记录时间        _ = session.run(target) #执行每次迭代        duration = time.time() - start_time        if i >= num_steps_burn_in:            if not i %10:                print('%s:step %d,duration = %.3f' %                      (datetime.now(),i-num_steps_burn_in,duration))            total_duration +=duration            total_duration_squared +=duration * duration #以便计算后面每轮耗时的均值和标准差            mn = total_duration / num_batches #平均耗时            vr = total_duration_squared / num_batches -mn * mn            sd = math.sqrt(vr) #标准差            print('%s:%s across %d steps,%.3f +/- %.3f sec /batch' % (datetime.now(),info_string,num_batches,mn,sd))batch_size = 32height,width=224,224inputs = tf.random_uniform((batch_size,height,width,3))with slim.arg_scope(resnet_arg_scope(is_training=False)):    net,end_points=resnet_v2_152(inputs,1000)init = tf.global_variables_initializer()sess = tf.Session()sess.run(init)num_batches=100time_tensorflow_run(sess,net,'Forward')