tensorflow卷積神經(jīng)Inception?V3網(wǎng)絡(luò)結(jié)構(gòu)代碼解析

更新時(shí)間：2022年05月06日 14:04:06 作者：fxfviolet

這篇文章主要為大家介紹了卷積神經(jīng)Inception?V3網(wǎng)絡(luò)結(jié)構(gòu)代碼解析，有需要的朋友可以借鑒參考下，希望能夠有所幫助，祝大家多多進(jìn)步，早日升職加薪

前言

學(xué)習(xí)了Inception V3卷積神經(jīng)網(wǎng)絡(luò)，總結(jié)一下對(duì)Inception V3網(wǎng)絡(luò)結(jié)構(gòu)和主要代碼的理解。

GoogLeNet對(duì)網(wǎng)絡(luò)中的傳統(tǒng)卷積層進(jìn)行了修改，提出了被稱為 Inception 的結(jié)構(gòu)，用于增加網(wǎng)絡(luò)深度和寬度，提高深度神經(jīng)網(wǎng)絡(luò)性能。從Inception V1到Inception V4有4個(gè)更新版本，每一版的網(wǎng)絡(luò)在原來的基礎(chǔ)上進(jìn)行改進(jìn)，提高網(wǎng)絡(luò)性能。本文介紹Inception V3的網(wǎng)絡(luò)結(jié)構(gòu)和主要代碼。

1 非Inception Module的普通卷積層

首先定義一個(gè)非Inception Module的普通卷積層函數(shù)inception_v3_base，輸入?yún)?shù)inputs為圖片數(shù)據(jù)的張量。第1個(gè)卷積層的輸出通道數(shù)為32，卷積核尺寸為【3x3】，步長為2，padding模式是默認(rèn)的VALID，第1個(gè)卷積層之后的張量尺寸變?yōu)?299-3)/2+1=149，即【149x149x32】。

后面的卷積層采用相同的形式，最后張量尺寸變?yōu)椤?5x35x192】。這幾個(gè)普通的卷積層主要使用了3x3的小卷積核，小卷積核可以低成本的跨通道的對(duì)特征進(jìn)行組合。

def inception_v3_base(inputs,scepe=None):
    with tf.variable_scope(scope,'InceptionV3',[inputs]):
        with slim.arg_scope([slim.conv2d,slim.max_pool2d,slim.avg_pool2d],stride=1,padding='VALID'):            
            # 149 x 149 x 32   
            net = slim.conv2d(inputs,32,[3,3],stride=2,scope='Conv2d_1a_3x3') 
            # 147 x 147 x 32'
            net = slim.conv2d(net,32),[3,3],scope='Conv2d_2a_3x3') 
            # 147 x 147 x 64
            net = slim.conv2d(net,64,[3,3],padding='SAME',scope='Conv2d_2b_3x3')  
            # 73 x 73 x 64
            net = slim.max_pool2d(net, [3, 3], stride=2, scope='MaxPool_3a_3x3')    
            # 73 x 73 x 80 
            net = slim.conv2d(net, 80, [1, 1], scope= 'Conv2d_3b_1x1')      
            # 71 x 71 x 192.
            net = slim.conv2d(net, 192, [3, 3], scope='Conv2d_4a_3x3',reuse=tf.AUTO_REUSE)    
            # 35 x 35 x 192
            net = slim.max_pool2d(net, [3, 3], stride=2, scope= 'MaxPool_5a_3x3')

2 三個(gè)Inception模塊組

接下來是三個(gè)連續(xù)的Inception模塊組，每個(gè)模塊組有多個(gè)Inception module組成。

下面是第1個(gè)Inception模塊組，包含了3個(gè)類似的Inception module，分別是：Mixed_5b，Mixed_5c，Mixed_5d。第1個(gè)Inception module有4個(gè)分支，

第1個(gè)分支是輸出通道為64的【1x1】卷積，

第2個(gè)分支是輸出通道為48的【1x1】卷積，再連接輸出通道為64的【5x5】卷積，

第3個(gè)分支是輸出通道為64的【1x1】卷積，再連接2個(gè)輸出通道為96的【3x3】卷積，

第4個(gè)分支是【3x3】的平均池化，再連接輸出通道為32的【1x1】卷積。

最后用tf.concat將4個(gè)分支的輸出合并在一起，輸出通道之和為54+64+96+32=256，最后輸出的張量尺寸為【35x35x256】。

第2個(gè)Inception module也有4個(gè)分支，與第1個(gè)模塊類似，只是最后連接輸出通道數(shù)為64的【1x1】卷積，最后輸出的張量尺寸為【35x35x288】。

第3個(gè)模塊與第2個(gè)模塊一樣。

    with slim.arg_scope([slim.conv2d,slim.max_pool2d,slim.avg_pool2d],stride=1,padding='SAME'):
        # 35 x 35 x 256
        end_point = 'Mixed_5b'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net,depth(64),[1,1],scope='Conv2d_0a_1x1')               
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(48), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(64), [5, 5], scope='Conv2d_0b_5x5')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(64), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(96), [3, 3],scope='Conv2d_0b_3x3')
                branch_2 = slim.conv2d(branch_2, depth(96), [3, 3], scope='Conv2d_0c_3x3')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(32), [1, 1], scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3]) # 64+64+96+32=256
        end_points[end_point] = net
        # 35 x 35 x 288
        end_point = 'Mixed_5c'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(64), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(48), [1, 1], scope='Conv2d_0b_1x1')
                branch_1 = slim.conv2d(branch_1, depth(64), [5, 5],scope='Conv_1_0c_5x5')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(64), [1, 1],scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(96), [3, 3],scope='Conv2d_0b_3x3')
                branch_2 = slim.conv2d(branch_2, depth(96), [3, 3],scope='Conv2d_0c_3x3')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3],scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(64), [1, 1],scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net
        # 35 x 35 x 288
        end_point = 'Mixed_5d'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(64), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(48), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(64), [5, 5],scope='Conv2d_0b_5x5')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(64), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(96), [3, 3],scope='Conv2d_0b_3x3')
                branch_2 = slim.conv2d(branch_2, depth(96), [3, 3],scope='Conv2d_0c_3x3')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(64), [1, 1],scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net

第2個(gè)Inception模塊組包含了5個(gè)Inception module，分別是Mixed_6a，Mixed_6b，Mixed_6ac，Mixed_6d，Mixed_6e。

每個(gè)Inception module包含有多個(gè)分支，第1個(gè)Inception module的步長為2，因此圖片尺寸被壓縮，最后輸出的張量尺寸為【17x17x768】。

第2個(gè)Inception module采用了Fractorization into small convolutions思想，串聯(lián)了【1x7】和【7x1】卷積，最后也是將多個(gè)通道合并。

第3、4個(gè)Inception module與第2個(gè)類似，都是用來增加卷積和非線性變化，提煉特征。張量尺寸不變，多個(gè)module后仍舊是【17x17x768】。

        # 17 x 17 x 768.
        end_point = 'Mixed_6a'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(384), [3, 3], stride=2,padding='VALID', scope='Conv2d_1a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(64), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(96), [3, 3],scope='Conv2d_0b_3x3')
                branch_1 = slim.conv2d(branch_1, depth(96), [3, 3], stride=2,padding='VALID', scope='Conv2d_1a_1x1')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID',scope='MaxPool_1a_3x3')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2]) # (35-3)/2+1=17
        end_points[end_point] = net
        # 17 x 17 x 768.
        end_point = 'Mixed_6b'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(128), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(128), [1, 7],scope='Conv2d_0b_1x7')
                branch_1 = slim.conv2d(branch_1, depth(192), [7, 1],scope='Conv2d_0c_7x1')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(128), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(128), [7, 1],scope='Conv2d_0b_7x1')
                branch_2 = slim.conv2d(branch_2, depth(128), [1, 7],scope='Conv2d_0c_1x7')
                branch_2 = slim.conv2d(branch_2, depth(128), [7, 1], scope='Conv2d_0d_7x1')
                branch_2 = slim.conv2d(branch_2, depth(192), [1, 7],scope='Conv2d_0e_1x7')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(192), [1, 1],scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net
        print(net.shape)
        # 17 x 17 x 768.
        end_point = 'Mixed_6c'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                ranch_0 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(160), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(160), [1, 7],scope='Conv2d_0b_1x7')
                branch_1 = slim.conv2d(branch_1, depth(192), [7, 1],scope='Conv2d_0c_7x1')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(160), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(160), [7, 1],scope='Conv2d_0b_7x1')
                branch_2 = slim.conv2d(branch_2, depth(160), [1, 7],scope='Conv2d_0c_1x7')
                branch_2 = slim.conv2d(branch_2, depth(160), [7, 1],scope='Conv2d_0d_7x1')
                branch_2 = slim.conv2d(branch_2, depth(192), [1, 7],scope='Conv2d_0e_1x7')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(192), [1, 1],scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net
        # 17 x 17 x 768.
        end_point = 'Mixed_6d'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(160), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(160), [1, 7], scope='Conv2d_0b_1x7')
                branch_1 = slim.conv2d(branch_1, depth(192), [7, 1], scope='Conv2d_0c_7x1')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(160), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(160), [7, 1], scope='Conv2d_0b_7x1')
                branch_2 = slim.conv2d(branch_2, depth(160), [1, 7], scope='Conv2d_0c_1x7')
                branch_2 = slim.conv2d(branch_2, depth(160), [7, 1], scope='Conv2d_0d_7x1')
                branch_2 = slim.conv2d(branch_2, depth(192), [1, 7], scope='Conv2d_0e_1x7')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], sco e='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(192), [1, 1],scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net
        # 17 x 17 x 768.
        end_point = 'Mixed_6e'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(192), [1, 7],
                                     scope='Conv2d_0b_1x7')
                branch_1 = slim.conv2d(branch_1, depth(192), [7, 1],
                                     scope='Conv2d_0c_7x1')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(branch_2, depth(192), [7, 1],
                                     scope='Conv2d_0b_7x1')
                branch_2 = slim.conv2d(branch_2, depth(192), [1, 7],
                                     scope='Conv2d_0c_1x7')
                branch_2 = slim.conv2d(branch_2, depth(192), [7, 1],
                                     scope='Conv2d_0d_7x1')
                branch_2 = slim.conv2d(branch_2, depth(192), [1, 7],
                                     scope='Conv2d_0e_1x7')
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(branch_3, depth(192), [1, 1],
                                     scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net

第3個(gè)Inception模塊組包含了3個(gè)Inception module，分別是Mxied_7a，Mixed_7b，Mixed_7c。

第1個(gè)Inception module包含了3個(gè)分支，與上面的結(jié)構(gòu)類似，主要也是通過改變通道數(shù)、卷積核尺寸，包括【1x1】、【3x3】、【1x7】、【7x1】來增加卷積和非線性變化，提升網(wǎng)絡(luò)性能。

最后3個(gè)分支在輸出通道上合并，輸出張量的尺寸為【8 x 8 x 1280】。第3個(gè)Inception module后得到的張量尺寸為【8 x 8 x 2048】。

        # 8 x 8 x 1280.
        end_point = 'Mixed_7a'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
                branch_0 = slim.conv2d(branch_0, depth(320), [3, 3], stride=2,
                                     padding='VALID', scope='Conv2d_1a_3x3')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(192), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = slim.conv2d(branch_1, depth(192), [1, 7],
                                     scope='Conv2d_0b_1x7')
                branch_1 = slim.conv2d(branch_1, depth(192), [7, 1],
                                     scope='Conv2d_0c_7x1')
                branch_1 = slim.conv2d(branch_1, depth(192), [3, 3], stride=2,
                                     padding='VALID', scope='Conv2d_1a_3x3')
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID',
                                         scope='MaxPool_1a_3x3')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2])
        end_points[end_point] = net
        # 8 x 8 x 2048.
        end_point = 'Mixed_7b'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(320), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(384), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = tf.concat(axis=3, values=[
                  slim.conv2d(branch_1, depth(384), [1, 3], scope='Conv2d_0b_1x3'),
                  slim.conv2d(branch_1, depth(384), [3, 1], scope='Conv2d_0b_3x1')])
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(448), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(
                  branch_2, depth(384), [3, 3], scope='Conv2d_0b_3x3')
                branch_2 = tf.concat(axis=3, values=[
                  slim.conv2d(branch_2, depth(384), [1, 3], scope='Conv2d_0c_1x3'),
                  slim.conv2d(branch_2, depth(384), [3, 1], scope='Conv2d_0d_3x1')])
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(
                  branch_3, depth(192), [1, 1], scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net)
        # 8 x 8 x 2048.
        end_point = 'Mixed_7c'
        with tf.variable_scope(end_point):
            with tf.variable_scope('Branch_0'):
                branch_0 = slim.conv2d(net, depth(320), [1, 1], scope='Conv2d_0a_1x1')
            with tf.variable_scope('Branch_1'):
                branch_1 = slim.conv2d(net, depth(384), [1, 1], scope='Conv2d_0a_1x1')
                branch_1 = tf.concat(axis=3, values=[
                  slim.conv2d(branch_1, depth(384), [1, 3], scope='Conv2d_0b_1x3'),
                  slim.conv2d(branch_1, depth(384), [3, 1], scope='Conv2d_0c_3x1')])
            with tf.variable_scope('Branch_2'):
                branch_2 = slim.conv2d(net, depth(448), [1, 1], scope='Conv2d_0a_1x1')
                branch_2 = slim.conv2d(
                  branch_2, depth(384), [3, 3], scope='Conv2d_0b_3x3')
                branch_2 = tf.concat(axis=3, values=[
                  slim.conv2d(branch_2, depth(384), [1, 3], scope='Conv2d_0c_1x3'),
                  slim.conv2d(branch_2, depth(384), [3, 1], scope='Conv2d_0d_3x1')])
            with tf.variable_scope('Branch_3'):
                branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')
                branch_3 = slim.conv2d(
                  branch_3, depth(192), [1, 1], scope='Conv2d_0b_1x1')
            net = tf.concat(axis=3, values=[branch_0, branch_1, branch_2, branch_3])
        end_points[end_point] = net

3 Auxiliary Logits、全局平均池化、Softmax分類

Inception V3網(wǎng)絡(luò)的最后一部分是Auxiliary Logits、全局平均池化、Softmax分類。

首先是Auxiliary Logits，作為輔助分類的節(jié)點(diǎn)，對(duì)分類結(jié)果預(yù)測(cè)有很大幫助。

先通過end_points['Mixed_6e']得到Mixed_6e后的特征張量，之后接一個(gè)【5x5】的平均池化，步長為3，padding為VALID，張量尺寸從第2個(gè)模塊組的【17x17x768】變?yōu)椤?x5x768】。

接著連接一個(gè)輸出通道為128的【1x1】卷積和輸出通道為768的【5x5】卷積，輸出尺寸變?yōu)椤?x1x768】。

然后連接輸出通道數(shù)為num_classes的【1x1】卷積，輸出變?yōu)椤?x1x1000】。最后將輔助分類節(jié)點(diǎn)的輸出存儲(chǔ)到字典表end_points中。

       with slim.arg_scope([slim.conv2d,slim.max_pool2d,slim.avg_pool2d],stride=1,padding='SAME'):
            aux_logits = end_points['Mixed_6e']
            print(aux_logits.shape)
            with tf.variable_scope('AuxLogits'):
                aux_logits = slim.avg_pool2d(aux_logits,[5,5],stride=3,padding='VALID',scope='AvgPool_1a_5x5')
                aux_logits = slim.conv2d(aux_logits,depth(128),[1,1],scope='Conv2d_1b_1x1')  # (17-5)/3+1=5
            kernel_size = _reduced_kernel_size_for_small_input(aux_logits, [5, 5])
            aux_logits = slim.conv2d(aux_logits, depth(768), kernel_size, weights_initializer=trunc_normal(0.01),
                                     padding='VALID', scope='Conv2d_2a_{}x{}'.format(*kernel_size))
            aux_logits = slim.conv2d( aux_logits, num_classes, [1, 1], activation_fn=None,
                                      normalizer_fn=None, weights_initializer=trunc_normal(0.001),
                                      scope='Conv2d_2b_1x1')         
            aux_logits = tf.squeeze(aux_logits, [1, 2], name='SpatialSqueeze')
            end_points['AuxLogits'] = aux_logits

最后對(duì)最后一個(gè)卷積層的輸出Mixed_7c進(jìn)行一個(gè)【8x8】的全局平均池化，padding為VALID，輸出張量從【8 x 8 x 2048】變?yōu)椤? x 1 x 2048】，然后連接一個(gè)Dropout層，接著連接一個(gè)輸出通道數(shù)為1000的【1x1】卷積。

使用tf.squeeze去掉輸出張量中維數(shù)為1的維度。最后用Softmax得到最終分類結(jié)果。返回分類結(jié)果logits和包含各個(gè)卷積后的特征圖字典表end_points。

        with tf.variable_scope('Logits'):
            kernel_size = _reduced_kernel_size_for_small_input(net, [8, 8])
            net = slim.avg_pool2d(net, kernel_size, padding='VALID',scope='AvgPool_1a_{}x{}'.format(*kernel_size))
            end_points['AvgPool_1a'] = net
            net = slim.dropout(net, keep_prob=dropout_keep_prob, scope='Dropout_1b')
            end_points['PreLogits'] = net 
            logits = slim.conv2d(net, num_classes, [1, 1], activation_fn=None, normalizer_fn=None, scope='Conv2d_1c_1x1')
            logits = tf.squeeze(logits, [1, 2], name='SpatialSqueeze')
            end_points['Logits'] = logits
            end_points['Predictions'] = slim.softmax(logits, scope='Predictions')
  return logits,end_points

參考文獻(xiàn)：

1. 《TensorFlow實(shí)戰(zhàn)》

以上就是卷積神經(jīng)Inception V3網(wǎng)絡(luò)結(jié)構(gòu)代碼解析的詳細(xì)內(nèi)容，更多關(guān)于Inception V3卷積神經(jīng)網(wǎng)絡(luò)的資料請(qǐng)關(guān)注腳本之家其它相關(guān)文章！

您可能感興趣的文章: