.. _sec_googlenet:
含并行连结的网络（GoogLeNet）
=============================
在2014年的ImageNet图像识别挑战赛中，一个名叫\ *GoogLeNet*
:cite:`Szegedy.Liu.Jia.ea.2015`\ 的网络架构大放异彩。
GoogLeNet吸收了NiN中串联网络的思想，并在此基础上做了改进。
这篇论文的一个重点是解决了什么样大小的卷积核最合适的问题。
毕竟，以前流行的网络使用小到\ :math:`1 \times 1`\ ，大到\ :math:`11 \times 11`\ 的卷积核。
本文的一个观点是，有时使用不同大小的卷积核组合是有利的。
本节将介绍一个稍微简化的GoogLeNet版本：我们省略了一些为稳定训练而添加的特殊特性，现在有了更好的训练方法，这些特性不是必要的。
Inception块
-----------
在GoogLeNet中，基本的卷积块被称为\ *Inception块*\ （Inception
block）。这很可能得名于电影《盗梦空间》（Inception），因为电影中的一句话“我们需要走得更深”（“We
need to go deeper”）。
.. _fig_inception:
.. figure:: ../img/inception.svg
Inception块的架构。
如 :numref:`fig_inception`\ 所示，Inception块由四条并行路径组成。
前三条路径使用窗口大小为\ :math:`1\times 1`\ 、\ :math:`3\times 3`\ 和\ :math:`5\times 5`\ 的卷积层，从不同空间大小中提取信息。
中间的两条路径在输入上执行\ :math:`1\times 1`\ 卷积，以减少通道数，从而降低模型的复杂性。
第四条路径使用\ :math:`3\times 3`\ 最大汇聚层，然后使用\ :math:`1\times 1`\ 卷积层来改变通道数。
这四条路径都使用合适的填充来使输入与输出的高和宽一致，最后我们将每条线路的输出在通道维度上连结，并构成Inception块的输出。在Inception块中，通常调整的超参数是每层输出通道数。
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
from mxnet import np, npx
from mxnet.gluon import nn
from d2l import mxnet as d2l
npx.set_np()
class Inception(nn.Block):
# c1--c4是每条路径的输出通道数
def __init__(self, c1, c2, c3, c4, **kwargs):
super(Inception, self).__init__(**kwargs)
# 线路1，单1x1卷积层
self.p1_1 = nn.Conv2D(c1, kernel_size=1, activation='relu')
# 线路2，1x1卷积层后接3x3卷积层
self.p2_1 = nn.Conv2D(c2[0], kernel_size=1, activation='relu')
self.p2_2 = nn.Conv2D(c2[1], kernel_size=3, padding=1,
activation='relu')
# 线路3，1x1卷积层后接5x5卷积层
self.p3_1 = nn.Conv2D(c3[0], kernel_size=1, activation='relu')
self.p3_2 = nn.Conv2D(c3[1], kernel_size=5, padding=2,
activation='relu')
# 线路4，3x3最大汇聚层后接1x1卷积层
self.p4_1 = nn.MaxPool2D(pool_size=3, strides=1, padding=1)
self.p4_2 = nn.Conv2D(c4, kernel_size=1, activation='relu')
def forward(self, x):
p1 = self.p1_1(x)
p2 = self.p2_2(self.p2_1(x))
p3 = self.p3_2(self.p3_1(x))
p4 = self.p4_2(self.p4_1(x))
# 在通道维度上连结输出
return np.concatenate((p1, p2, p3, p4), axis=1)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l
class Inception(nn.Module):
# c1--c4是每条路径的输出通道数
def __init__(self, in_channels, c1, c2, c3, c4, **kwargs):
super(Inception, self).__init__(**kwargs)
# 线路1，单1x1卷积层
self.p1_1 = nn.Conv2d(in_channels, c1, kernel_size=1)
# 线路2，1x1卷积层后接3x3卷积层
self.p2_1 = nn.Conv2d(in_channels, c2[0], kernel_size=1)
self.p2_2 = nn.Conv2d(c2[0], c2[1], kernel_size=3, padding=1)
# 线路3，1x1卷积层后接5x5卷积层
self.p3_1 = nn.Conv2d(in_channels, c3[0], kernel_size=1)
self.p3_2 = nn.Conv2d(c3[0], c3[1], kernel_size=5, padding=2)
# 线路4，3x3最大汇聚层后接1x1卷积层
self.p4_1 = nn.MaxPool2d(kernel_size=3, stride=1, padding=1)
self.p4_2 = nn.Conv2d(in_channels, c4, kernel_size=1)
def forward(self, x):
p1 = F.relu(self.p1_1(x))
p2 = F.relu(self.p2_2(F.relu(self.p2_1(x))))
p3 = F.relu(self.p3_2(F.relu(self.p3_1(x))))
p4 = F.relu(self.p4_2(self.p4_1(x)))
# 在通道维度上连结输出
return torch.cat((p1, p2, p3, p4), dim=1)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
import tensorflow as tf
from d2l import tensorflow as d2l
class Inception(tf.keras.Model):
# c1--c4是每条路径的输出通道数
def __init__(self, c1, c2, c3, c4):
super().__init__()
# 线路1，单1x1卷积层
self.p1_1 = tf.keras.layers.Conv2D(c1, 1, activation='relu')
# 线路2，1x1卷积层后接3x3卷积层
self.p2_1 = tf.keras.layers.Conv2D(c2[0], 1, activation='relu')
self.p2_2 = tf.keras.layers.Conv2D(c2[1], 3, padding='same',
activation='relu')
# 线路3，1x1卷积层后接5x5卷积层
self.p3_1 = tf.keras.layers.Conv2D(c3[0], 1, activation='relu')
self.p3_2 = tf.keras.layers.Conv2D(c3[1], 5, padding='same',
activation='relu')
# 线路4，3x3最大汇聚层后接1x1卷积层
self.p4_1 = tf.keras.layers.MaxPool2D(3, 1, padding='same')
self.p4_2 = tf.keras.layers.Conv2D(c4, 1, activation='relu')
def call(self, x):
p1 = self.p1_1(x)
p2 = self.p2_2(self.p2_1(x))
p3 = self.p3_2(self.p3_1(x))
p4 = self.p4_2(self.p4_1(x))
# 在通道维度上连结输出
return tf.keras.layers.Concatenate()([p1, p2, p3, p4])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
import warnings
from d2l import paddle as d2l
warnings.filterwarnings("ignore")
import paddle
import paddle.nn as nn
import paddle.nn.functional as F
class Inception(nn.Layer):
# c1--c4是每条路径的输出通道数
def __init__(self, in_channels, c1, c2, c3, c4, **kwargs):
super(Inception, self).__init__(**kwargs)
# 线路1，单1x1卷积层
self.p1_1 = nn.Conv2D(in_channels, c1, kernel_size=1)
# 线路2，1x1卷积层后接3x3卷积层
self.p2_1 = nn.Conv2D(in_channels, c2[0], kernel_size=1)
self.p2_2 = nn.Conv2D(c2[0], c2[1], kernel_size=3, padding=1)
# 线路3，1x1卷积层后接5x5卷积层
self.p3_1 = nn.Conv2D(in_channels, c3[0], kernel_size=1)
self.p3_2 = nn.Conv2D(c3[0], c3[1], kernel_size=5, padding=2)
# 线路4，3x3最大池化层后接1x1卷积层
self.p4_1 = nn.MaxPool2D(kernel_size=3, stride=1, padding=1)
self.p4_2 = nn.Conv2D(in_channels, c4, kernel_size=1)
def forward(self, x):
p1 = F.relu(self.p1_1(x))
p2 = F.relu(self.p2_2(F.relu(self.p2_1(x))))
p3 = F.relu(self.p3_2(F.relu(self.p3_1(x))))
p4 = F.relu(self.p4_2(self.p4_1(x)))
# 在通道维度上连结输出
return paddle.concat(x=[p1, p2, p3, p4], axis=1)
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b1 = nn.Sequential()
b1.add(nn.Conv2D(64, kernel_size=7, strides=2, padding=3, activation='relu'),
nn.MaxPool2D(pool_size=3, strides=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b1 = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3),
nn.ReLU(),
nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def b1():
return tf.keras.models.Sequential([
tf.keras.layers.Conv2D(64, 7, strides=2, padding='same',
activation='relu'),
tf.keras.layers.MaxPool2D(pool_size=3, strides=2, padding='same')])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b1 = nn.Sequential(nn.Conv2D(1, 64, kernel_size=7, stride=2, padding=3),
nn.ReLU(),
nn.MaxPool2D(kernel_size=3, stride=2,padding=1))
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
W0818 09:46:08.883092 94744 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.8, Runtime API Version: 11.8
W0818 09:46:08.925873 94744 gpu_resources.cc:91] device: 0, cuDNN Version: 8.7.
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b2 = nn.Sequential()
b2.add(nn.Conv2D(64, kernel_size=1, activation='relu'),
nn.Conv2D(192, kernel_size=3, padding=1, activation='relu'),
nn.MaxPool2D(pool_size=3, strides=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b2 = nn.Sequential(nn.Conv2d(64, 64, kernel_size=1),
nn.ReLU(),
nn.Conv2d(64, 192, kernel_size=3, padding=1),
nn.ReLU(),
nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def b2():
return tf.keras.Sequential([
tf.keras.layers.Conv2D(64, 1, activation='relu'),
tf.keras.layers.Conv2D(192, 3, padding='same', activation='relu'),
tf.keras.layers.MaxPool2D(pool_size=3, strides=2, padding='same')])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b2 = nn.Sequential(nn.Conv2D(64, 64, kernel_size=1),
nn.ReLU(),
nn.Conv2D(64, 192, kernel_size=3, padding=1),
nn.ReLU(),
nn.MaxPool2D(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b3 = nn.Sequential()
b3.add(Inception(64, (96, 128), (16, 32), 32),
Inception(128, (128, 192), (32, 96), 64),
nn.MaxPool2D(pool_size=3, strides=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b3 = nn.Sequential(Inception(192, 64, (96, 128), (16, 32), 32),
Inception(256, 128, (128, 192), (32, 96), 64),
nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def b3():
return tf.keras.models.Sequential([
Inception(64, (96, 128), (16, 32), 32),
Inception(128, (128, 192), (32, 96), 64),
tf.keras.layers.MaxPool2D(pool_size=3, strides=2, padding='same')])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b3 = nn.Sequential(Inception(192, 64, (96, 128), (16, 32), 32),
Inception(256, 128, (128, 192), (32, 96), 64),
nn.MaxPool2D(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b4 = nn.Sequential()
b4.add(Inception(192, (96, 208), (16, 48), 64),
Inception(160, (112, 224), (24, 64), 64),
Inception(128, (128, 256), (24, 64), 64),
Inception(112, (144, 288), (32, 64), 64),
Inception(256, (160, 320), (32, 128), 128),
nn.MaxPool2D(pool_size=3, strides=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b4 = nn.Sequential(Inception(480, 192, (96, 208), (16, 48), 64),
Inception(512, 160, (112, 224), (24, 64), 64),
Inception(512, 128, (128, 256), (24, 64), 64),
Inception(512, 112, (144, 288), (32, 64), 64),
Inception(528, 256, (160, 320), (32, 128), 128),
nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def b4():
return tf.keras.Sequential([
Inception(192, (96, 208), (16, 48), 64),
Inception(160, (112, 224), (24, 64), 64),
Inception(128, (128, 256), (24, 64), 64),
Inception(112, (144, 288), (32, 64), 64),
Inception(256, (160, 320), (32, 128), 128),
tf.keras.layers.MaxPool2D(pool_size=3, strides=2, padding='same')])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b4 = nn.Sequential(Inception(480, 192, (96, 208), (16, 48), 64),
Inception(512, 160, (112, 224), (24, 64), 64),
Inception(512, 128, (128, 256), (24, 64), 64),
Inception(512, 112, (144, 288), (32, 64), 64),
Inception(528, 256, (160, 320), (32, 128), 128),
nn.MaxPool2D(kernel_size=3, stride=2, padding=1))
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b5 = nn.Sequential()
b5.add(Inception(256, (160, 320), (32, 128), 128),
Inception(384, (192, 384), (48, 128), 128),
nn.GlobalAvgPool2D())
net = nn.Sequential()
net.add(b1, b2, b3, b4, b5, nn.Dense(10))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b5 = nn.Sequential(Inception(832, 256, (160, 320), (32, 128), 128),
Inception(832, 384, (192, 384), (48, 128), 128),
nn.AdaptiveAvgPool2d((1,1)),
nn.Flatten())
net = nn.Sequential(b1, b2, b3, b4, b5, nn.Linear(1024, 10))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def b5():
return tf.keras.Sequential([
Inception(256, (160, 320), (32, 128), 128),
Inception(384, (192, 384), (48, 128), 128),
tf.keras.layers.GlobalAvgPool2D(),
tf.keras.layers.Flatten()
])
# “net”必须是一个将被传递给“d2l.train_ch6（）”的函数。
# 为了利用我们现有的CPU/GPU设备，这样模型构建/编译需要在“strategy.scope()”
def net():
return tf.keras.Sequential([b1(), b2(), b3(), b4(), b5(),
tf.keras.layers.Dense(10)])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
b5 = nn.Sequential(Inception(832, 256, (160, 320), (32, 128), 128),
Inception(832, 384, (192, 384), (48, 128), 128),
nn.AdaptiveAvgPool2D((1, 1)),
nn.Flatten())
net = nn.Sequential(b1, b2, b3, b4, b5, nn.Linear(1024, 10))
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = np.random.uniform(size=(1, 1, 96, 96))
net.initialize()
for layer in net:
X = layer(X)
print(layer.name, 'output shape:\t', X.shape)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
[07:34:43] ../src/storage/storage.cc:196: Using Pooled (Naive) StorageManager for CPU
sequential0 output shape: (1, 64, 24, 24)
sequential1 output shape: (1, 192, 12, 12)
sequential2 output shape: (1, 480, 6, 6)
sequential3 output shape: (1, 832, 3, 3)
sequential4 output shape: (1, 1024, 1, 1)
dense0 output shape: (1, 10)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = torch.rand(size=(1, 1, 96, 96))
for layer in net:
X = layer(X)
print(layer.__class__.__name__,'output shape:\t', X.shape)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
Sequential output shape: torch.Size([1, 64, 24, 24])
Sequential output shape: torch.Size([1, 192, 12, 12])
Sequential output shape: torch.Size([1, 480, 6, 6])
Sequential output shape: torch.Size([1, 832, 3, 3])
Sequential output shape: torch.Size([1, 1024])
Linear output shape: torch.Size([1, 10])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = tf.random.uniform(shape=(1, 96, 96, 1))
for layer in net().layers:
X = layer(X)
print(layer.__class__.__name__, 'output shape:\t', X.shape)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
Sequential output shape: (1, 24, 24, 64)
Sequential output shape: (1, 12, 12, 192)
Sequential output shape: (1, 6, 6, 480)
Sequential output shape: (1, 3, 3, 832)
Sequential output shape: (1, 1024)
Dense output shape: (1, 10)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = paddle.rand(shape=(1, 1, 96, 96))
for layer in net:
X = layer(X)
print(layer.__class__.__name__,'output shape:\t', X.shape)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
Sequential output shape: [1, 64, 24, 24]
Sequential output shape: [1, 192, 12, 12]
Sequential output shape: [1, 480, 6, 6]
Sequential output shape: [1, 832, 3, 3]
Sequential output shape: [1, 1024]
Linear output shape: [1, 10]
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
lr, num_epochs, batch_size = 0.1, 10, 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
loss 0.256, train acc 0.903, test acc 0.907
2343.6 examples/sec on gpu(0)
.. figure:: output_googlenet_83a8b4_108_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
lr, num_epochs, batch_size = 0.1, 10, 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
loss 0.262, train acc 0.900, test acc 0.886
3265.5 examples/sec on cuda:0
.. figure:: output_googlenet_83a8b4_111_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
lr, num_epochs, batch_size = 0.1, 10, 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
loss 0.248, train acc 0.905, test acc 0.899
3758.1 examples/sec on /GPU:0
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
.. figure:: output_googlenet_83a8b4_114_2.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
lr, num_epochs, batch_size = 0.1, 10, 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
loss 0.213, train acc 0.919, test acc 0.904
1381.9 examples/sec on Place(gpu:0)
.. figure:: output_googlenet_83a8b4_117_1.svg
.. raw:: html
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html