本站内容均来自兴趣收集,如不慎侵害的您的相关权益,请留言告知,我们将尽快删除.谢谢.
持续创作,加速成长!这是我参与「掘金日新计划 · 6 月更文挑战」的第11天,点击查看活动详情
:tangerine:作者简介:秃头小苏,致力于用最通俗的语言描述问题
:tangerine:往期回顾: 基于pytorch搭建VGGNet神经网络用于花类识别 基于pytorch搭建AlexNet神经网络用于花类识别
:tangerine:近期目标:拥有5000粉丝
:tangerine:支持小苏:点赞:+1|type_3:、收藏:star:、留言:envelope_with_arrow:
基于pytorch搭建ResNet神经网络用于花类识别
写在前面
这一系列已经写了好几篇了,这篇结束后可能就会停更一系列了,因为一方面,看懂了已经更新的这些我认为其他的网络大概就是照葫芦画瓢,自己多多少少是能看明白个大概的。 【当然这是要在你对这部分网络结构的理论有充分的了解之后】 另一方面,我觉得这部分真的得你自己切切实实的钻研,自己一步步的调试,看别人的文章、甚至是视频,你可能会得到短暂的满足,但是许多细节你是体验不到的。所以这里给出 基于pytorch搭建ResNet神经网络用于花类识别 的完整代码,希望大家下去后仔细阅读:rose::rose::rose:
至于这一系列再次更新的话不出意外会讲讲一些轻量级网络像MobileNet、shuffleNet等,当然了这部分都已经做过理论部分的概述了:lemon::lemon::lemon:
还是回归到本文上来,首先你需要具备以下知识:
ResNet的理论
pytorch搭建网络基础
当然,这些我在前文都已经介绍过,大家抱着缺啥补啥的态度去看看呗:ear_of_rice::ear_of_rice::ear_of_rice:
深度学习经典网络模型汇总
使用pytorch自己构建网络模型实战
基于pytorch搭建AlexNet神经网络用于花类识别
基于pytorch搭建VGGNet神经网络用于花类识别
ResNet网络模型搭建:sparkles::sparkles::sparkles:
自己写文章的宗旨是致力于用最通俗的语言描述问题嘛:dart::dart::dart:但是对于一些关乎于代码的文章,有的时候单纯的文字确实很难将问题表述清楚,因此我建议你先观看此视频,对ResNet网络模型搭建的整理流程有了一个大致的了解后再来阅读此文,然后再以这篇文章为辅进行学习,这样我觉得是高效的学习方式:four_leaf_clover::four_leaf_clover::four_leaf_clover:【 当然这样还是不够的,你一定要自己去独立的阅读代码,一步步的调试运行,这一点我想我再强调也不为过】
ResNeta网络是有大量重复的结构堆叠而成的,它的网络层数主要有18层、34层、50层、101层和152层。对于18层和34层的网络它的基础模块为 basic block ,而对于50层、101层和152层的网络其基础模块为 bottleneck block 。我们可以分别来定义这两个基础模块,如下:
# 定义BasicBlock class BasicBlock(nn.Module): expansion = 1 def __init__(self, in_channel, out_channel, stride=1, downsample=None, **kwargs): super(BasicBlock, self).__init__() self.conv1 = nn.Conv2d(in_channels=in_channel, out_channels=out_channel, kernel_size=3, stride=stride, padding=1, bias=False) # 特征图尺寸不变 self.bn1 = nn.BatchNorm2d(out_channel) # BN层建议放在卷积和激活层之间 self.relu = nn.ReLU() self.conv2 = nn.Conv2d(in_channels=out_channel, out_channels=out_channel, kernel_size=3, stride=1, padding=1, bias=False) self.bn2 = nn.BatchNorm2d(out_channel) self.downsample = downsample def forward(self, x): identity = x if self.downsample is not None: identity = self.downsample(x) out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) out += identity out = self.relu(out) return out
# 定义Bottleneck class Bottleneck(nn.Module): """ 注意:原论文中,在虚线残差结构的主分支上,第一个1x1卷积层的步距是2,第二个3x3卷积层步距是1。 但在pytorch官方实现过程中是第一个1x1卷积层的步距是1,第二个3x3卷积层步距是2, 这幺做的好处是能够在top1上提升大概0.5%的准确率。 可参考Resnet v1.5 https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch """ expansion = 4 def __init__(self, in_channel, out_channel, stride=1, downsample=None, groups=1, width_per_group=64): super(Bottleneck, self).__init__() width = int(out_channel * (width_per_group / 64.)) * groups self.conv1 = nn.Conv2d(in_channels=in_channel, out_channels=width, kernel_size=1, stride=1, bias=False) # squeeze channels self.bn1 = nn.BatchNorm2d(width) # ----------------------------------------- self.conv2 = nn.Conv2d(in_channels=width, out_channels=width, groups=groups, kernel_size=3, stride=stride, bias=False, padding=1) self.bn2 = nn.BatchNorm2d(width) # ----------------------------------------- self.conv3 = nn.Conv2d(in_channels=width, out_channels=out_channel*self.expansion, kernel_size=1, stride=1, bias=False) # unsqueeze channels self.bn3 = nn.BatchNorm2d(out_channel*self.expansion) self.relu = nn.ReLU(inplace=True) self.downsample = downsample def forward(self, x): identity = x if self.downsample is not None: identity = self.downsample(x) out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) out = self.relu(out) out = self.conv3(out) out = self.bn3(out) out += identity out = self.relu(out) return out
接着我们就可以来定义我们的ResNet网络了:
class ResNet(nn.Module): def __init__(self, block, blocks_num, num_classes=1000, include_top=True, groups=1, width_per_group=64): super(ResNet, self).__init__() self.include_top = include_top self.in_channel = 64 self.groups = groups self.width_per_group = width_per_group self.conv1 = nn.Conv2d(3, self.in_channel, kernel_size=7, stride=2, padding=3, bias=False) self.bn1 = nn.BatchNorm2d(self.in_channel) self.relu = nn.ReLU(inplace=True) self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) self.layer1 = self._make_layer(block, 64, blocks_num[0]) self.layer2 = self._make_layer(block, 128, blocks_num[1], stride=2) self.layer3 = self._make_layer(block, 256, blocks_num[2], stride=2) self.layer4 = self._make_layer(block, 512, blocks_num[3], stride=2) if self.include_top: self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) # output size = (1, 1) self.fc = nn.Linear(512 * block.expansion, num_classes) for m in self.modules(): if isinstance(m, nn.Conv2d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') def forward(self, x): x = self.conv1(x) x = self.bn1(x) x = self.relu(x) x = self.maxpool(x) x = self.layer1(x) x = self.layer2(x) x = self.layer3(x) x = self.layer4(x) if self.include_top: x = self.avgpool(x) x = torch.flatten(x, 1) x = self.fc(x) return x
我们可以看出再ResNet 的定义中有这样的函数:
该函数表示对ResNet的每个基础模块一个整合,即layer1对应conv2_x、layer2对应conv3_x、layer3对应conv4_x、layer4对应conv5_x:rice::rice::rice:
_make_layer
函数的定义如下:
def _make_layer(self, block, channel, block_num, stride=1): downsample = None if stride != 1 or self.in_channel != channel * block.expansion: downsample = nn.Sequential( nn.Conv2d(self.in_channel, channel * block.expansion, kernel_size=1, stride=stride, bias=False), nn.BatchNorm2d(channel * block.expansion)) layers = [] layers.append(block(self.in_channel, channel, downsample=downsample, stride=stride, groups=self.groups, width_per_group=self.width_per_group)) self.in_channel = channel * block.expansion for _ in range(1, block_num): layers.append(block(self.in_channel, channel, groups=self.groups, width_per_group=self.width_per_group)) return nn.Sequential(*layers) def forward(self, x): x = self.conv1(x) x = self.bn1(x) x = self.relu(x) x = self.maxpool(x) x = self.layer1(x) x = self.layer2(x) x = self.layer3(x) x = self.layer4(x) if self.include_top: x = self.avgpool(x) x = torch.flatten(x, 1) x = self.fc(x) return x
最后我们来看看如何定义一个具体的网络:
def resnet34(num_classes=1000, include_top=True): # https://download.pytorch.org/models/resnet34-333f7ec4.pth return ResNet(BasicBlock, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top) def resnet50(num_classes=1000, include_top=True): # https://download.pytorch.org/models/resnet50-19c8e357.pth return ResNet(Bottleneck, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top) def resnet101(num_classes=1000, include_top=True): # https://download.pytorch.org/models/resnet101-5d3b4d8f.pth return ResNet(Bottleneck, [3, 4, 23, 3], num_classes=num_classes, include_top=include_top)
训练结构展示
ResNet34训练结果:
ResNet50训练结果:
ResNet101训练结果:
迁移学习使用ResNet34预加载模型:
下面给出各种模型生成的权重文件,如下:
小结
这一部分我感到有一些的奇怪,即上文用resnet训时,resnet101和resnet50的效果要比resnet34效果差,但是理论部分不是说resnet层数深效果越好嘛,这是什幺原因呢?希望知道的可以告知。:herb::herb::herb:
问了一些大佬,对于上述问题他们的解答是:这个和自己任务也有关系,简单的任务可能用小网络效果更好。
参考视频: www.bilibili.com/video/BV14E…
如若文章对你有所帮助,那就
Be First to Comment