pytorch使用之微調網絡模型

import torchvision.models as models

1.調整最後一層輸出維度

model = models.ResNet(pretrained=True)
fc_features = model.fc.in_features# 獲取全連接層輸入維度
model.fc = torch.nn.Linear(fc_features, num_class)
``

2.調整某一層參數

import torch
import torch.nn as nn
from torchvision import models

class ft_net(nn.Module):
    def __init__(self, class_num = 751):
        super(ft_net, self).__init__()
        #load the model
        model_ft = models.resnet50(pretrained=True) 
        if stride == 1:
        	model_ft.layer4[0].downsample[0].stride = (1,1)
        	model_ft.layer4[0].conv2.stride = (1,1)
        model_ft.avgpool = nn.AdaptiveAvgPool2d((1,1))
        self.model = model_ft
        self.fc = nn.Linear(2048, num_class)
    def forward(self, x):
        x = self.model.conv1(x)
        x = self.model.bn1(x)
        x = self.model.relu(x)
        x = self.model.maxpool(x)
        x = self.model.layer1(x)
        x = self.model.layer2(x)
        x = self.model.layer3(x)
        x = self.model.layer4(x)
        x = self.model.avgpool(x)
        x = torch.squeeze(x)
        x = self.fc(x) #use our classifier.
        return x

調整模型中layer4[0]中的downsample[0]的stride值。

添加層並加載參數

import torchvision.models as models
import torch
import torch.nn as nn
import math
import torch.utils.model_zoo as model_zoo
 
class CNN(nn.Module):
    def __init__(self, block, layers, num_classes=9):
        self.inplanes = 64
        super(ResNet, self).__init__()
        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
                               bias=False)
        self.bn1 = nn.BatchNorm2d(64)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, 64, layers[0])
        self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
        self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
        self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
        self.avgpool = nn.AvgPool2d(7, stride=1)
        #新增一個反捲積層
        self.convtranspose1 = nn.ConvTranspose2d(2048, 2048, kernel_size=3, stride=1, padding=1, output_padding=0, groups=1, bias=False, dilation=1)
        #新增一個最大池化層
        self.maxpool2 = nn.MaxPool2d(kernel_size=3, stride=1, padding=1)
        #去掉原來的fc層，新增一個fclass層
        self.fclass = nn.Linear(2048, num_classes)
 
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
                m.weight.data.normal_(0, math.sqrt(2. / n))
            elif isinstance(m, nn.BatchNorm2d):
                m.weight.data.fill_(1)
                m.bias.data.zero_()
 
    def _make_layer(self, block, planes, blocks, stride=1):
        downsample = None
        if stride != 1 or self.inplanes != planes * block.expansion:
            downsample = nn.Sequential(
                nn.Conv2d(self.inplanes, planes * block.expansion,
                          kernel_size=1, stride=stride, bias=False),
                nn.BatchNorm2d(planes * block.expansion),
            )
 
        layers = []
        layers.append(block(self.inplanes, planes, stride, downsample))
        self.inplanes = planes * block.expansion
        for i in range(1, blocks):
            layers.append(block(self.inplanes, planes))
 
        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)
 
        x = self.avgpool(x)
        #新加層的forward
        x = x.view(x.size(0), -1)
        x = self.convtranspose1(x)
        x = self.maxpool2(x)
        x = x.view(x.size(0), -1)
        x = self.fclass(x)
 
        return x
#加載model
resnet50 = models.resnet50(pretrained=True)
cnn = CNN(Bottleneck, [3, 4, 6, 3])
#讀取參數
pretrained_dict = resnet50.state_dict()
model_dict = cnn.state_dict()
#將pretrained_dict裏不屬於model_dict的鍵剔除掉
pretrained_dict =  {k: v for k, v in pretrained_dict.items() if k in model_dict}
#更新現有的model_dict
model_dict.update(pretrained_dict)
#加載我們真正需要的state_dict
cnn.load_state_dict(model_dict)
#print(resnet50)
print(cnn)

多GPU訓練參數模型加載：
使用多GPU訓練權重需要使用model = torch.nn.Parallel(model)處理模型

對於參數模型中可能包含state_dict,epoch,accruccy,等鍵值對，我們只需要加載key爲state_dict的鍵值對。
例：

pretrained_dict = torch.load(dict_path)
model_dict_clone = pretrained_dict.copy()
for key, value in model_dict_clone.items():
	if key == ‘state_dict’:
		pretrained_dict = model_dict_clone['state_dict']

單個GPU訓練使用多GPU加載
由於單個GPU訓練保存的權重參數，key沒有module前綴，使用並行GPU不能直接加載模型參數。需要進行鍵值對的key進行處理,需要用list將字典包裝，否則在迭代處理字典的時候會出現錯誤

pretrained_dict = torch.load(dict_path)
model_dict_clone = pretrained_dict.copy()
for key, value in list(model_dict_clone.items()):
	model_dict_clone['module.' + key] = model_dict_clone.pop(key)