厦门网站改版,网络营销的成功案例,互联网做什么比较赚钱,哪里有网络课程平台网站_就是帮老师建设一个教学的网站目标 知道GoogLeNet网络结构的特点能够利用GoogLeNet完成图像分类 一、开发背景 GoogLeNet在2014年由Google团队提出#xff0c; 斩获当年ImageNet(ILSVRC14)竞赛中Classification Task (分类任务) 第一名#xff0c;VGG获得了第二名#xff0c;为了向“LeNet”致敬#x…目标 知道GoogLeNet网络结构的特点能够利用GoogLeNet完成图像分类 一、开发背景 GoogLeNet在2014年由Google团队提出 斩获当年ImageNet(ILSVRC14)竞赛中Classification Task (分类任务) 第一名VGG获得了第二名为了向“LeNet”致敬因此取名为“GoogLeNet”。 GoogLeNet做了更加大胆的网络结构尝试虽然深度只有22层但大小却比AlexNet和VGG小很多。GoogleNet参数为500万个AlexNet参数个数是GoogleNet的12倍VGGNet参数又是AlexNet的3倍因此在内存或计算资源有限时GoogleNet是比较好的选择从模型结果来看GoogLeNet的性能也更加优越。 GoogLeNet的名字不是GoogleNet而是GoogLeNet这是为了致敬LeNet。GoogLeNet和AlexNet/VGGNet这类依靠加深网络结构的深度的思想不完全一样。GoogLeNet在加深度的同时做了结构上的创新引入了一个叫做Inception的结构来代替之前的卷积加激活的经典组件。GoogLeNet在ImageNet分类比赛上的Top-5错误率降低到了6.7%。 1.Inception 块 GoogLeNet中的基础卷积块叫作Inception块得名于同名电影《盗梦空间》Inception。Inception块在结构比较复杂如下图所示 Inception块里有4条并行的线路。前3条线路使用窗口大小分别是1×11×1、3×33×3和5×55×5的卷积层来抽取不同空间尺寸下的信息其中中间2个线路会对输入先做1×11×1卷积来减少输入通道数以降低模型复杂度。第4条线路则使用3×33×3最大池化层后接1×11×1卷积层来改变通道数。4条线路都使用了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上连结,并向后进行传输。 1×11×1卷积 它的计算方法和其他卷积核一样唯一不同的是它的大小是1×11×1没有考虑在特征图局部信息之间的关系。 它的作用主要是 实现跨通道的交互和信息整合 卷积核通道数的降维和升维减少网络参数 在tf.keras中实现Inception模块各个卷积层卷积核的个数通过输入参数来控制如下所示 # 定义Inception模块 class Inception(tf.keras.layers.Layer):# 输入参数为各个卷积的卷积核个数def __init__(self, c1, c2, c3, c4):super().__init__()# 线路11 x 1卷积层激活函数是RELUpadding是sameself.p1_1 tf.keras.layers.Conv2D(c1, kernel_size1, activationrelu, paddingsame)# 线路21 x 1卷积层后接3 x 3卷积层,激活函数是RELUpadding是sameself.p2_1 tf.keras.layers.Conv2D(c2[0], kernel_size1, paddingsame, activationrelu)self.p2_2 tf.keras.layers.Conv2D(c2[1], kernel_size3, paddingsame,activationrelu)# 线路31 x 1卷积层后接5 x 5卷积层,激活函数是RELUpadding是sameself.p3_1 tf.keras.layers.Conv2D(c3[0], kernel_size1, paddingsame, activationrelu)self.p3_2 tf.keras.layers.Conv2D(c3[1], kernel_size5, paddingsame,activationrelu)# 线路43 x 3最大池化层后接1 x 1卷积层,激活函数是RELUpadding是sameself.p4_1 tf.keras.layers.MaxPool2D(pool_size3, paddingsame, strides1)self.p4_2 tf.keras.layers.Conv2D(c4, kernel_size1, paddingsame, activationrelu)# 完成前向传播过程def call(self, x):# 线路1p1 self.p1_1(x)# 线路2p2 self.p2_2(self.p2_1(x))# 线路3p3 self.p3_2(self.p3_1(x))# 线路4p4 self.p4_2(self.p4_1(x))# 在通道维上concat输出outputs tf.concat([p1, p2, p3, p4], axis-1)return outputs 指定通道数对Inception模块进行实例化 Inception(64, (96, 128), (16, 32), 32) 2.GoogLeNet模型 GoogLeNet主要由Inception模块构成如下图所示 整个网络架构我们分为五个模块每个模块之间使用步幅为2的3×33×3最大池化层来减小输出高宽。 2.1 B1模块 第一模块使用一个64通道的7×77×7卷积层。 # 定义模型的输入 inputs tf.keras.Input(shape(224,224,3),name input) # b1 模块 # 卷积层7*7的卷积核步长为2pad是same激活函数RELU x tf.keras.layers.Conv2D(64, kernel_size7, strides2, paddingsame, activationrelu)(inputs) # 最大池化窗口大小为3*3步长为2pad是same x tf.keras.layers.MaxPool2D(pool_size3, strides2, paddingsame)(x) # b2 模块 2.2 B2模块 第二模块使用2个卷积层首先是64通道的1×11×1卷积层然后是将通道增大3倍的3×33×3卷积层。 # b2 模块 # 卷积层1*1的卷积核步长为2pad是same激活函数RELU x tf.keras.layers.Conv2D(64, kernel_size1, paddingsame, activationrelu)(x) # 卷积层3*3的卷积核步长为2pad是same激活函数RELU x tf.keras.layers.Conv2D(192, kernel_size3, paddingsame, activationrelu)(x) # 最大池化窗口大小为3*3步长为2pad是same x tf.keras.layers.MaxPool2D(pool_size3, strides2, paddingsame)(x) 2.3 B3模块 第三模块串联2个完整的Inception块。第一个Inception块的输出通道数为641283232256641283232256。第二个Inception块输出通道数增至1281929664480 # b3 模块 # Inception x Inception(64, (96, 128), (16, 32), 32)(x) # Inception x Inception(128, (128, 192), (32, 96), 64)(x) # 最大池化窗口大小为3*3步长为2pad是same x tf.keras.layers.MaxPool2D(pool_size3, strides2, paddingsame)(x) 2.4 B4模块 第四模块更加复杂。它串联了5个Inception块其输出通道数分别是19220848645121922084864512、16022464645121602246464512、12825664645121282566464512、11228864645281122886464528和256320128128832256320128128832。并且增加了辅助分类器根据实验发现网络的中间层具有很强的识别能力为了利用中间层抽象的特征在某些中间层中添加含有多层的分类器如下图所示 实现如下所示 def aux_classifier(x, filter_size):#x:输入数据filter_size:卷积层卷积核个数全连接层神经元个数# 池化层x tf.keras.layers.AveragePooling2D(pool_size5, strides3, paddingsame)(x)# 1x1 卷积层x tf.keras.layers.Conv2D(filtersfilter_size[0], kernel_size1, strides1,paddingvalid, activationrelu)(x)# 展平x tf.keras.layers.Flatten()(x)# 全连接层1x tf.keras.layers.Dense(unitsfilter_size[1], activationrelu)(x)# softmax输出层x tf.keras.layers.Dense(units10, activationsoftmax)(x)return x b4模块的实现 # b4 模块 # Inception x Inception(192, (96, 208), (16, 48), 64)(x) # 辅助输出1 aux_output_1 aux_classifier(x, [128, 1024]) # Inception x Inception(160, (112, 224), (24, 64), 64)(x) # Inception x Inception(128, (128, 256), (24, 64), 64)(x) # Inception x Inception(112, (144, 288), (32, 64), 64)(x) # 辅助输出2 aux_output_2 aux_classifier(x, [128, 1024]) # Inception x Inception(256, (160, 320), (32, 128), 128)(x) # 最大池化 x tf.keras.layers.MaxPool2D(pool_size3, strides2, paddingsame)(x) 2.5 B5模块 第五模块有输出通道数为256320128128832256320128128832和38438412812810243843841281281024的两个Inception块。后面紧跟输出层该模块使用全局平均池化层GAP来将每个通道的高和宽变成1。最后输出变成二维数组后接输出个数为标签类别数的全连接层。 全局平均池化层GAP 用来替代全连接层将特征图每一通道中所有像素值相加后求平均得到就是GAP的结果在将其送入后续网络中进行计算 实现过程是 # b5 模块 # Inception x Inception(256, (160, 320), (32, 128), 128)(x) # Inception x Inception(384, (192, 384), (48, 128), 128)(x) # GAP x tf.keras.layers.GlobalAvgPool2D()(x) # 输出层 main_outputs tf.keras.layers.Dense(10,activationsoftmax)(x) # 使用Model来创建模型指明输入和输出 构建GoogLeNet模型并通过summary来看下模型的结构 # 使用Model来创建模型指明输入和输出 model tf.keras.Model(inputsinputs, outputs[main_outputs,aux_output_1aux_output_2]) model.summary() Model: functional_3 _________________________________________________________________ Layer (type) Output Shape Param # input (InputLayer) [(None, 224, 224, 3)] 0 _________________________________________________________________ conv2d_122 (Conv2D) (None, 112, 112, 64) 9472 _________________________________________________________________ max_pooling2d_27 (MaxPooling (None, 56, 56, 64) 0 _________________________________________________________________ conv2d_123 (Conv2D) (None, 56, 56, 64) 4160 _________________________________________________________________ conv2d_124 (Conv2D) (None, 56, 56, 192) 110784 _________________________________________________________________ max_pooling2d_28 (MaxPooling (None, 28, 28, 192) 0 _________________________________________________________________ inception_19 (Inception) (None, 28, 28, 256) 163696 _________________________________________________________________ inception_20 (Inception) (None, 28, 28, 480) 388736 _________________________________________________________________ max_pooling2d_31 (MaxPooling (None, 14, 14, 480) 0 _________________________________________________________________ inception_21 (Inception) (None, 14, 14, 512) 376176 _________________________________________________________________ inception_22 (Inception) (None, 14, 14, 512) 449160 _________________________________________________________________ inception_23 (Inception) (None, 14, 14, 512) 510104 _________________________________________________________________ inception_24 (Inception) (None, 14, 14, 528) 605376 _________________________________________________________________ inception_25 (Inception) (None, 14, 14, 832) 868352 _________________________________________________________________ max_pooling2d_37 (MaxPooling (None, 7, 7, 832) 0 _________________________________________________________________ inception_26 (Inception) (None, 7, 7, 832) 1043456 _________________________________________________________________ inception_27 (Inception) (None, 7, 7, 1024) 1444080 _________________________________________________________________ global_average_pooling2d_2 ( (None, 1024) 0 _________________________________________________________________ dense_10 (Dense) (None, 10) 10250 Total params: 5,983,802 Trainable params: 5,983,802 Non-trainable params: 0 ___________________________________________________________ 3.手写数字识别 因为ImageNet数据集较大训练时间较长我们仍用前面的MNIST数据集来演示GoogLeNet。读取数据的时将图像高和宽扩大到图像高和宽224。这个通过tf.image.resize_with_pad来实现。 2.1 数据读取 首先获取数据,并进行维度调整 import numpy as np # 获取手写数字数据集 (train_images, train_labels), (test_images, test_labels) mnist.load_data() # 训练集数据维度的调整N H W C train_images np.reshape(train_images,(train_images.shape[0],train_images.shape[1],train_images.shape[2],1)) # 测试集数据维度的调整N H W C test_images np.reshape(test_images,(test_images.shape[0],test_images.shape[1],test_images.shape[2],1)) 由于使用全部数据训练时间较长我们定义两个方法获取部分数据并将图像调整为224*224大小进行模型训练(与VGG中是一样的) # 定义两个方法随机抽取部分样本演示 # 获取训练集数据 def get_train(size):# 随机生成要抽样的样本的索引index np.random.randint(0, np.shape(train_images)[0], size)# 将这些数据resize成22*227大小resized_images tf.image.resize_with_pad(train_images[index],224,224,)# 返回抽取的return resized_images.numpy(), train_labels[index] # 获取测试集数据 def get_test(size):# 随机生成要抽样的样本的索引index np.random.randint(0, np.shape(test_images)[0], size)# 将这些数据resize成224*224大小resized_images tf.image.resize_with_pad(test_images[index],224,224,)# 返回抽样的测试样本return resized_images.numpy(), test_labels[index] 调用上述两个方法获取参与模型训练和测试的数据集 # 获取训练样本和测试样本 train_images,train_labels get_train(256) test_images,test_labels get_test(128) 3.2 模型编译 # 指定优化器损失函数和评价指标 optimizer tf.keras.optimizers.SGD(learning_rate0.01, momentum0.0) # 模型有3个输出所以指定损失函数对应的权重系数 net.compile(optimizeroptimizer,losssparse_categorical_crossentropy,metrics[accuracy],loss_weights[1,0.3,0.3]) 3.3 模型训练 # 模型训练指定训练数据batchsize,epoch,验证集 net.fit(train_images,train_labels,batch_size128,epochs3,verbose1,validation_split0.1) 训练过程 Epoch 1/3 2/2 [] - 8s 4s/step - loss: 2.9527 - accuracy: 0.1174 - val_loss: 3.3254 - val_accuracy: 0.1154 Epoch 2/3 2/2 [] - 7s 4s/step - loss: 2.8111 - accuracy: 0.0957 - val_loss: 2.2718 - val_accuracy: 0.2308 Epoch 3/3 2/2 [] - 7s 4s/step - loss: 2.3055 - accuracy: 0.0957 - val_loss: 2.2669 - val_accuracy: 0.2308 2.4 模型评估 # 指定测试数据 net.evaluate(test_images,test_labels,verbose1) 输出为 4/4 [] - 1s 338ms/step - loss: 2.3110 - accuracy: 0.0781 [2.310971260070801, 0.078125] 4.延伸版本 GoogLeNet是以InceptionV1为基础进行构建的所以GoogLeNet也叫做InceptionNet,在随后的⼏年⾥研究⼈员对GoogLeNet进⾏了数次改进 就又产生了InceptionV2V3,V4等版本。 4.1 InceptionV2 在InceptionV2中将大卷积核拆分为小卷积核将V1中的5×55×5的卷积用两个3×33×3的卷积替代从而增加网络的深度减少了参数。 4.2 InceptionV3 将n×n卷积分割为1×n和n×1两个卷积例如一个的3×33×3卷积首先执行一个1×31×3的卷积然后执行一个3×13×1的卷积,这种方法的参数量和计算量都比原来降低。 总结 知道GoogLeNet的网络架构有基础模块Inception构成能够利用GoogleNet完成图像分类 下面是另一个代码实现GooLeNet网络模型构建和之前代码不冲突 GooLeNet代码实现 展示模型搭建代码 import torch import torch.nn as nn import torch.nn.functional as F#convReLU class BasicConv2d(nn.Module):def __init__(self, in_channels, out_channels, **kwargs):super(BasicConv2d, self).__init__()self.conv nn.Conv2d(in_channels, out_channels, **kwargs)self.relu nn.ReLU()def forward(self, x):x self.conv(x)x self.relu(x)return x#前部 class Front(nn.Module):def __init__(self):super(Front, self).__init__()self.conv1 BasicConv2d(3, 64, kernel_size7, stride2, padding3)self.maxpool1 nn.MaxPool2d(3, stride2,ceil_modeTrue)self.conv2 BasicConv2d(64, 64, kernel_size1)self.conv3 BasicConv2d(64, 192, kernel_size3, padding1)self.maxpool2 nn.MaxPool2d(3, stride2,ceil_modeTrue)def forward(self,input):#输入(N,3,224,224)x self.conv1(input)#(N,64,112,112)x self.maxpool1(x)#(N,64,56,56)x self.conv2(x)#(N,64,56,56)x self.conv3(x)#(N,192,56,56)x self.maxpool2(x)#(N,192,28,28)return xclass Inception(nn.Module):def __init__(self, in_channels, ch1x1, ch3x3_1_1, ch3x3_1, ch3x3_2_1, ch3x3_2, pool_ch):super(Inception, self).__init__()self.branch1 BasicConv2d(in_channels, ch1x1, kernel_size1)self.branch2 nn.Sequential(BasicConv2d(in_channels, ch3x3_1_1, kernel_size1),BasicConv2d(ch3x3_1_1, ch3x3_1, kernel_size3, padding1))self.branch3 nn.Sequential(BasicConv2d(in_channels, ch3x3_2_1, kernel_size1),BasicConv2d(ch3x3_2_1, ch3x3_2, kernel_size3, padding1))self.branch4 nn.Sequential(nn.MaxPool2d(kernel_size3, stride1, padding1),BasicConv2d(in_channels, pool_ch, kernel_size1))def forward(self, x):#输入(N,Cin,Hin,Win)branch1 self.branch1(x)#(N,C1,Hin,Win)branch2 self.branch2(x)#(N,C2,Hin,Win)branch3 self.branch3(x)#(N,C3,Hin,Win)branch4 self.branch4(x)#(N,C4,Hin,Win)outputs [branch1, branch2, branch3, branch4]return torch.cat(outputs, 1)#(N,C1C2C3C4,Hin,Win)#辅助分类器 class InceptionAux(nn.Module):def __init__(self, in_channels, num_classes):super(InceptionAux, self).__init__()self.averagePool nn.AvgPool2d(kernel_size5, stride3)self.conv BasicConv2d(in_channels, 128, kernel_size1)self.fc1 nn.Linear(2048, 1024)self.fc2 nn.Linear(1024, num_classes)def forward(self, x):# 输入aux1:(N,512,14,14), aux2: (N,528,14,14)x self.averagePool(x)# aux1:(N,512,4,4), aux2: (N,528,4,4)x self.conv(x)# (N,128,4,4)x torch.flatten(x, 1)# (N,2048)x F.dropout(x, 0.5, trainingself.training)x F.relu(self.fc1(x))# (N,1024)x F.dropout(x, 0.5, trainingself.training)x self.fc2(x)# (N,num_classes)return x# GooLeNet网络主体 class GoogLeNet(nn.Module):def __init__(self, num_classes1000, aux_logitsTrue):super(GoogLeNet, self).__init__()self.aux_logits aux_logitsself.front Front()self.inception3a Inception(192, 64, 96, 128, 16, 32, 32)self.inception3b Inception(256, 128, 128, 192, 32, 96, 64)self.maxpool3 nn.MaxPool2d(3, stride2,ceil_modeTrue)self.inception4a Inception(480, 192, 96, 208, 16, 48, 64)self.inception4b Inception(512, 160, 112, 224, 24, 64, 64)self.inception4c Inception(512, 128, 128, 256, 24, 64, 64)self.inception4d Inception(512, 112, 144, 288, 32, 64, 64)self.inception4e Inception(528, 256, 160, 320, 32, 128, 128)self.maxpool4 nn.MaxPool2d(3, stride2,ceil_modeTrue)self.inception5a Inception(832, 256, 160, 320, 32, 128, 128)self.inception5b Inception(832, 384, 192, 384, 48, 128, 128)if self.aux_logits:self.aux1 InceptionAux(512, num_classes)self.aux2 InceptionAux(528, num_classes)self.avgpool nn.AdaptiveAvgPool2d((1, 1))self.dropout nn.Dropout(0.4)self.fc nn.Linear(1024, num_classes)def forward(self, x):#输入(N,3,224,224)x self.front(x)#(N,192,28,28)x self.inception3a(x)#(N,256,28,28)x self.inception3b(x)#(N,480,28,28)x self.maxpool3(x)#(N,480,14,14)x self.inception4a(x)#(N,512,14,14)if self.training and self.aux_logits:aux1 self.aux1(x)x self.inception4b(x)#(N,512,14,14)x self.inception4c(x)#(N,512,14,14)x self.inception4d(x)#(N,528,14,14)if self.training and self.aux_logits:aux2 self.aux2(x)x self.inception4e(x)#(N,832,14,14)x self.maxpool4(x)#(N,832,7,7)x self.inception5a(x)#(N,832,7,7)x self.inception5b(x)#(N,1024,7,7)x self.avgpool(x)#(N,1024,1,1)x torch.flatten(x, 1)#(N,1024)x self.dropout(x)x self.fc(x)#(N,num_classes)if self.training and self.aux_logits:return x, aux2, aux1return x 使用 Pytorch 搭建 GoogleNet 网络 本代码使用的数据集来自 “花分类” 数据集→ 传送门 ←具体内容看 data_set文件夹下的 README.md model.py 搭建 GoogleNet 网络模型 import torch.nn as nn import torch import torch.nn.functional as Fclass GoogleNet(nn.Module):# aux_logits: 是否使用辅助分类器训练的时候为True, 验证的时候为False)def __init__(self, num_classes1000, aux_logitsTrue, init_weightFalse):super(GoogleNet, self).__init__()self.aux_logits aux_logitsself.conv1 BasicConv2d(3, 64, kernel_size7, stride2, padding3)self.maxpool1 nn.MaxPool2d(3, stride2, ceil_modeTrue) # 当结构为小数时ceil_modeTrue向上取整False向下取整# nn.LocalResponseNorm 此处省略self.conv2 nn.Sequential(BasicConv2d(64, 64, kernel_size1),BasicConv2d(64, 192, kernel_size3, padding1))self.maxpool2 nn.MaxPool2d(3, stride2, ceil_modeTrue)self.inception3a Inception(192, 64, 96, 128, 16, 32, 32)self.inception3b Inception(256, 128, 128, 192, 32, 96, 64)self.maxpool3 nn.MaxPool2d(3, stride2, ceil_modeTrue)self.inception4a Inception(480, 192, 96, 208, 16, 48, 64)self.inception4b Inception(512, 160, 112, 224, 24, 64, 64)self.inception4c Inception(512, 128, 128, 256, 24, 64, 64)self.inception4d Inception(512, 112, 144, 288, 32, 64, 64)self.inception4e Inception(528, 256, 160, 320, 32, 128, 128)self.maxpool4 nn.MaxPool2d(2, stride2, ceil_modeTrue)self.inception5a Inception(832, 256, 160, 320, 32, 128, 128)self.inception5b Inception(832, 384, 192, 384, 48, 128, 128)if aux_logits: # 使用辅助分类器self.aux1 InceptionAux(512, num_classes)self.aux2 InceptionAux(528, num_classes)self.avgpool nn.AdaptiveAvgPool1d((1, 1))self.dropout nn.Dropout(0.4)self.fc nn.Linear(1024, num_classes)if init_weight:self._initialize_weight()def forward(self, x):x self.conv1(x)x self.maxpool1(x)x self.conv2(x)x self.maxpool2(x)x self.inception3a(x)x self.inception3b(x)x self.maxpool3(x)x self.inception4a(x)if self.training and self.aux_logits:aux1 self.aux1(x)x self.inception4b(x)x self.inception4c(x)x self.inception4d(x)if self.training and self.aux_logits:aux2 self.aux2(x)x self.inception4e(x)x self.maxpool4(x)x self.inception5a(x)x self.inception5b(x)x self.avgpool(x)x torch.flatten(x, 1)x self.dropout(x)x self.fc(x)if self.training and self.aux_logits:return x, aux1, aux2return xdef _initialize_weight(self):for m in self.modules():if isinstance(m, nn.Conv2d):nn.init.kaiming_normal_(m.weight, modefan_out, nonlinearity)if m.bias is not None:nn.init.constant_(m.bias, 0)elif isinstance(m, nn.Linear):nn.init.normal_(m.weight, 0, 0.01)nn.init.constant_(m.bias, 0)# 创建 Inception 结构函数模板 class Inception(nn.Module):# 参数为 Inception 结构的那几个卷积核的数量详细见表def __init__(self, in_channels, ch1x1, ch3x3red, ch3x3, ch5x5red, ch5x5, pool_proj):super(Inception, self).__init__()# 四个并联结构self.branch1 BasicConv2d(in_channels, ch1x1, kernel_size1)self.branch2 nn.Sequential(BasicConv2d(in_channels, ch3x3red, kernel_size1),BasicConv2d(ch3x3red, ch3x3, kernel_size3, padding1))self.branch3 nn.Sequential(BasicConv2d(in_channels, ch5x5red, kernel_size1),BasicConv2d(ch5x5red, ch5x5, kernel_size5, padding2))self.branch4 nn.Sequential(nn.MaxPool2d(kernel_size3, stride1, padding1),BasicConv2d(in_channels, pool_proj, kernel_size1))def forward(self, x):branch1 self.branch1(x)branch2 self.branch2(x)branch3 self.branch3(x)branch4 self.branch4(x)outputs [branch1, branch2, branch3, branch4]return torch.cat(outputs, 1)# 创建辅助分类器结构函数模板 class InceptionAux(nn.Module):def __init__(self, in_channels, num_classes):super(InceptionAux, self).__init__()self.avgPool nn.AvgPool2d(kernel_size5, stride3)self.conv BasicConv2d(in_channels, 128, kernel_size1)self.fc1 nn.Linear(2048, 1024)self.fc2 nn.Linear(1024, num_classes)def forward(self, x):# aux1: N x 512 x 14 x 14 aux2: N x 528 x 14 x 14输入x self.avgPool(x)# aux1: N x 512 x 4 x 4 aux2: N x 528 x 4 x 4输出 4 (14 - 5)/3 1x self.conv(x)x torch.flatten(x, 1) # 展平x F.dropout(x, 0.5, trainingself.training)x F.relu(self.fc1(x), inplaceTrue)x F.dropout(x, 0.5, trainingself.training)x self.fc2(x)return x# 创建卷积层函数模板 class BasicConv2d(nn.Module):def __init__(self, in_channels, out_channels, **kwargs):super(BasicConv2d, self).__init__()self.conv nn.Conv2d(in_channels, out_channels, **kwargs)self.relu nn.ReLU(True)def forward(self, x):x self.conv(x)x self.relu(x)return x train.py 训练网络 import os import jsonimport torch import torch.nn as nn from torchvision import transforms, datasets import torch.optim as optim from tqdm import tqdmfrom model import GoogleNetdef main():device torch.device(cuda:0 if torch.cuda.is_available() else cpu)print(using {} device..format(device))data_transform {train: transforms.Compose([transforms.RandomResizedCrop(224),transforms.RandomHorizontalFlip(),transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))]),val: transforms.Compose([transforms.Resize((224, 224)),transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])}data_root os.path.abspath(os.path.join(os.getcwd(), ../..)) # get data root pathimage_path os.path.join(data_root, data_set, flower_data) # flower data set pathassert os.path.exists(image_path), {} path does not exist..format(image_path)train_dataset datasets.ImageFolder(rootos.path.join(image_path, train),transformdata_transform[train])train_num len(train_dataset)# {daisy:0, dandelion:1, roses:2, sunflower:3, tulips:4}flower_list train_dataset.class_to_idxcla_dict dict((val, key) for key, val in flower_list.items())# write dict into json filejson_str json.dumps(cla_dict, indent4)with open(class_indices.json, w) as json_file:json_file.write(json_str)batch_size 32nw min([os.cpu_count(), batch_size if batch_size 1 else 0, 8]) # number of workersprint(Using {} dataloader workers every process.format(nw))train_loader torch.utils.data.DataLoader(train_dataset,batch_sizebatch_size, shuffleTrue,num_workersnw)validate_dataset datasets.ImageFolder(rootos.path.join(image_path, val),transformdata_transform[val])val_num len(validate_dataset)validate_loader torch.utils.data.DataLoader(validate_dataset,batch_sizebatch_size, shuffleFalse,num_workersnw)print(using {} images for training, {} images for validation..format(train_num,val_num))net GoogleNet(num_classes5, aux_logitsTrue, init_weightsTrue)net.to(device)loss_function nn.CrossEntropyLoss()optimizer optim.Adam(net.parameters(), lr0.0003)epochs 30best_acc 0.0save_path ./googleNet.pthtrain_steps len(train_loader)for epoch in range(epochs):# trainnet.train()running_loss 0.0train_bar tqdm(train_loader)for step, data in enumerate(train_bar):images, labels dataoptimizer.zero_grad()logits, aux_logits2, aux_logits1 net(images.to(device)) # 由于训练的时候会使用辅助分类器所有相当于有三个返回结果loss0 loss_function(logits, labels.to(device))loss1 loss_function(aux_logits1, labels.to(device))loss2 loss_function(aux_logits2, labels.to(device))loss loss0 loss1 * 0.3 loss2 * 0.3loss.backward()optimizer.step()# print statisticsrunning_loss loss.item()train_bar.desc train epoch[{}/{}] loss:{:.3f}.format(epoch 1,epochs,loss)# validatenet.eval()acc 0.0 # accumulate accurate number / epochwith torch.no_grad():val_bar tqdm(validate_loader)for val_data in val_bar:val_images, val_labels val_dataoutputs net(val_images.to(device)) # eval model only have last output layerpredict_y torch.max(outputs, dim1)[1]acc torch.eq(predict_y, val_labels.to(device)).sum().item()val_accurate acc / val_numprint([epoch %d] train_loss: %.3f val_accuracy: %.3f %(epoch 1, running_loss / train_steps, val_accurate))if val_accurate best_acc:best_acc val_accuratetorch.save(net.state_dict(), save_path)print(Finished Training)if __name__ __main__:main()predict.py 使用训练好的模型网络对图像分类 import os import jsonimport torch from PIL import Image from torchvision import transforms import matplotlib.pyplot as pltfrom model import GoogleNetdef main():device torch.device(cuda:0 if torch.cuda.is_available() else cpu)data_transform transforms.Compose([transforms.Resize((224, 224)),transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])# load imageimg_path ../tulip.jpgassert os.path.exists(img_path), file: {} dose not exist..format(img_path)img Image.open(img_path)plt.imshow(img)# [N, C, H, W]img data_transform(img)# expand batch dimensionimg torch.unsqueeze(img, dim0)# read class_indictjson_path ./class_indices.jsonassert os.path.exists(json_path), file: {} dose not exist..format(json_path)json_file open(json_path, r)class_indict json.load(json_file)# create modelmodel GoogleNet(num_classes5, aux_logitsFalse).to(device)# load model weightsweights_path ./googleNet.pthassert os.path.exists(weights_path), file: {} dose not exist..format(weights_path)missing_keys, unexpected_keys model.load_state_dict(torch.load(weights_path, map_locationdevice),strictFalse)model.eval()with torch.no_grad():# predict classoutput torch.squeeze(model(img.to(device))).cpu()predict torch.softmax(output, dim0)predict_cla torch.argmax(predict).numpy()print_res class: {} prob: {:.3}.format(class_indict[str(predict_cla)],predict[predict_cla].numpy())plt.title(print_res)print(print_res)plt.show()if __name__ __main__:main()参考文章【学习笔记】GoogleNet 网络结构_googlenet特点-CSDN博客 参考文章GoogLeNet详解-CSDN博客 参考文章CNN经典网络模型四GoogLeNet简介及代码实现PyTorch超详细注释版_googlenet代码-CSDN博客