Pytorch模型迁移和迁移学习,导入部分模型参数的操作
1.利用resnet18做迁移学习
importtorch
fromtorchvisionimportmodels
if__name__=="__main__":
#device=torch.device("cuda"iftorch.cuda.is_available()else"cpu")
device='cpu'
print("-----device:{}".format(device))
print("-----Pytorchversion:{}".format(torch.__version__))
input_tensor=torch.zeros(1,3,100,100)
print('input_tensor:',input_tensor.shape)
pretrained_file="model/resnet18-5c106cde.pth"
model=models.resnet18()
model.load_state_dict(torch.load(pretrained_file))
model.eval()
out=model(input_tensor)
print("out:",out.shape,out[0,0:10])
结果输出:
input_tensor:torch.Size([1,3,100,100])
out:torch.Size([1,1000])tensor([0.4010,0.8436,0.3072,0.0627,0.4446,0.8470,0.1882,0.7012,0.2988,-0.7574],grad_fn=)
如果,我们修改了resnet18的网络结构,如何将原来预训练模型参数(resnet18-5c106cde.pth)迁移到新的resnet18网络中呢?
比如,这里将官方的resnet18的self.layer4=self._make_layer(block,512,layers[3],stride=2)改为:self.layer44=self._make_layer(block,512,layers[3],stride=2)
classResNet(nn.Module): def__init__(self,block,layers,num_classes=1000,zero_init_residual=False): super(ResNet,self).__init__() self.inplanes=64 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.layer44=self._make_layer(block,512,layers[3],stride=2) self.avgpool=nn.AdaptiveAvgPool2d((1,1)) self.fc=nn.Linear(512*block.expansion,num_classes) forminself.modules(): ifisinstance(m,nn.Conv2d): nn.init.kaiming_normal_(m.weight,mode='fan_out',nonlinearity='relu') elifisinstance(m,nn.BatchNorm2d): nn.init.constant_(m.weight,1) nn.init.constant_(m.bias,0) #Zero-initializethelastBNineachresidualbranch, #sothattheresidualbranchstartswithzeros,andeachresidualblockbehaveslikeanidentity. #Thisimprovesthemodelby0.2~0.3%accordingtohttps://arxiv.org/abs/1706.02677 ifzero_init_residual: forminself.modules(): ifisinstance(m,Bottleneck): nn.init.constant_(m.bn3.weight,0) elifisinstance(m,BasicBlock): nn.init.constant_(m.bn2.weight,0) def_make_layer(self,block,planes,blocks,stride=1): downsample=None ifstride!=1orself.inplanes!=planes*block.expansion: downsample=nn.Sequential( conv1x1(self.inplanes,planes*block.expansion,stride), nn.BatchNorm2d(planes*block.expansion), ) layers=[] layers.append(block(self.inplanes,planes,stride,downsample)) self.inplanes=planes*block.expansion for_inrange(1,blocks): layers.append(block(self.inplanes,planes)) returnnn.Sequential(*layers) defforward(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.layer44(x) x=self.avgpool(x) x=x.view(x.size(0),-1) x=self.fc(x) returnx
这时,直接加载模型:
model=models.resnet18() model.load_state_dict(torch.load(pretrained_file))
这时,肯定会报错,类似：Missingkey(s)instate_dict或者Unexpectedkey(s)instate_dict的错误：
RuntimeError:Error(s)inloadingstate_dictforResNet:
Missingkey(s)instate_dict:"layer44.0.conv1.weight","layer44.0.bn1.weight","layer44.0.bn1.bias","layer44.0.bn1.running_mean","layer44.0.bn1.running_var","layer44.0.conv2.weight","layer44.0.bn2.weight","layer44.0.bn2.bias","layer44.0.bn2.running_mean","layer44.0.bn2.running_var","layer44.0.downsample.0.weight","layer44.0.downsample.1.weight","layer44.0.downsample.1.bias","layer44.0.downsample.1.running_mean","layer44.0.downsample.1.running_var","layer44.1.conv1.weight","layer44.1.bn1.weight","layer44.1.bn1.bias","layer44.1.bn1.running_mean","layer44.1.bn1.running_var","layer44.1.conv2.weight","layer44.1.bn2.weight","layer44.1.bn2.bias","layer44.1.bn2.running_mean","layer44.1.bn2.running_var".
Unexpectedkey(s)instate_dict:"layer4.0.conv1.weight","layer4.0.bn1.running_mean","layer4.0.bn1.running_var","layer4.0.bn1.weight","layer4.0.bn1.bias","layer4.0.conv2.weight","layer4.0.bn2.running_mean","layer4.0.bn2.running_var","layer4.0.bn2.weight","layer4.0.bn2.bias","layer4.0.downsample.0.weight","layer4.0.downsample.1.running_mean","layer4.0.downsample.1.running_var","layer4.0.downsample.1.weight","layer4.0.downsample.1.bias","layer4.1.conv1.weight","layer4.1.bn1.running_mean","layer4.1.bn1.running_var","layer4.1.bn1.weight","layer4.1.bn1.bias","layer4.1.conv2.weight","layer4.1.bn2.running_mean","layer4.1.bn2.running_var","layer4.1.bn2.weight","layer4.1.bn2.bias".Processfinishedwith
RuntimeError:Error(s)inloadingstate_dictforResNet:
Unexpectedkey(s)instate_dict:"layer4.0.conv1.weight","layer4.0.bn1.running_mean","layer4.0.bn1.running_var","layer4.0.bn1.weight","layer4.0.bn1.bias","layer4.0.conv2.weight","layer4.0.bn2.running_mean","layer4.0.bn2.running_var","layer4.0.bn2.weight","layer4.0.bn2.bias","layer4.0.downsample.0.weight","layer4.0.downsample.1.running_mean","layer4.0.downsample.1.running_var","layer4.0.downsample.1.weight","layer4.0.downsample.1.bias","layer4.1.conv1.weight","layer4.1.bn1.running_mean","layer4.1.bn1.running_var","layer4.1.bn1.weight","layer4.1.bn1.bias","layer4.1.conv2.weight","layer4.1.bn2.running_mean","layer4.1.bn2.running_var","layer4.1.bn2.weight","layer4.1.bn2.bias".
我们希望将原来预训练模型参数(resnet18-5c106cde.pth)迁移到新的resnet18网络,当然只能迁移二者相同的模型参数,不同的参数还是随机初始化的.
deftransfer_model(pretrained_file,model):
'''
只导入pretrained_file部分模型参数
tensor([-0.7119,0.0688,-1.7247,-1.7182,-1.2161,-0.7323,-2.1065,-0.5433,-1.5893,-0.5562]
update:
D.update([E,]**F)->None.UpdateDfromdict/iterableEandF.
IfEispresentandhasa.keys()method,thendoes:forkinE:D[k]=E[k]
IfEispresentandlacksa.keys()method,thendoes:fork,vinE:D[k]=v
Ineithercase,thisisfollowedby:forkinF:D[k]=F[k]
:parampretrained_file:
:parammodel:
:return:
'''
pretrained_dict=torch.load(pretrained_file)#getpretraineddict
model_dict=model.state_dict()#getmodeldict
#在合并前(update),需要去除pretrained_dict一些不需要的参数
pretrained_dict=transfer_state_dict(pretrained_dict,model_dict)
model_dict.update(pretrained_dict)#更新(合并)模型的参数
model.load_state_dict(model_dict)
returnmodel
deftransfer_state_dict(pretrained_dict,model_dict):
'''
根据model_dict,去除pretrained_dict一些不需要的参数,以便迁移到新的网络
url:https://blog.csdn.net/qq_34914551/article/details/87871134
:parampretrained_dict:
:parammodel_dict:
:return:
'''
#state_dict2={k:vfork,vinsave_model.items()ifkinmodel_dict.keys()}
state_dict={}
fork,vinpretrained_dict.items():
ifkinmodel_dict.keys():
#state_dict.setdefault(k,v)
state_dict[k]=v
else:
print("Missingkey(s)instate_dict:{}".format(k))
returnstate_dict
if__name__=="__main__":
input_tensor=torch.zeros(1,3,100,100)
print('input_tensor:',input_tensor.shape)
pretrained_file="model/resnet18-5c106cde.pth"
#model=resnet18()
#model.load_state_dict(torch.load(pretrained_file))
#model.eval()
#out=model(input_tensor)
#print("out:",out.shape,out[0,0:10])
model1=resnet18()
model1=transfer_model(pretrained_file,model1)
out1=model1(input_tensor)
print("out1:",out1.shape,out1[0,0:10])
2.修改网络名称并迁移学习
上面的例子,只是将官方的resnet18的self.layer4=self._make_layer(block,512,layers[3],stride=2)改为了:self.layer44=self._make_layer(block,512,layers[3],stride=2),我们仅仅是修改了一个网络名称而已,就导致model.load_state_dict(torch.load(pretrained_file))出错,
那么,我们如何将预训练模型"model/resnet18-5c106cde.pth"转换成符合新的网络的模型参数呢?
方法很简单,只需要将resnet18-5c106cde.pth的模型参数中所有前缀为layer4的名称,改为layer44即可
本人已经定义好了方法:
modify_state_dict(pretrained_dict,model_dict,old_prefix,new_prefix)
defstring_rename(old_string,new_string,start,end):
new_string=old_string[:start]+new_string+old_string[end:]
returnnew_string
defmodify_model(pretrained_file,model,old_prefix,new_prefix):
'''
:parampretrained_file:
:parammodel:
:paramold_prefix:
:paramnew_prefix:
:return:
'''
pretrained_dict=torch.load(pretrained_file)
model_dict=model.state_dict()
state_dict=modify_state_dict(pretrained_dict,model_dict,old_prefix,new_prefix)
model.load_state_dict(state_dict)
returnmodel
defmodify_state_dict(pretrained_dict,model_dict,old_prefix,new_prefix):
'''
修改modeldict
:parampretrained_dict:
:parammodel_dict:
:paramold_prefix:
:paramnew_prefix:
:return:
'''
state_dict={}
fork,vinpretrained_dict.items():
ifkinmodel_dict.keys():
#state_dict.setdefault(k,v)
state_dict[k]=v
else:
foro,ninzip(old_prefix,new_prefix):
prefix=k[:len(o)]
ifprefix==o:
kk=string_rename(old_string=k,new_string=n,start=0,end=len(o))
print("renamelayermodules:{}-->{}".format(k,kk))
state_dict[kk]=v
returnstate_dict
if__name__=="__main__":
input_tensor=torch.zeros(1,3,100,100)
print('input_tensor:',input_tensor.shape)
pretrained_file="model/resnet18-5c106cde.pth"
#model=models.resnet18()
#model.load_state_dict(torch.load(pretrained_file))
#model.eval()
#out=model(input_tensor)
#print("out:",out.shape,out[0,0:10])
#
#model1=resnet18()
#model1=transfer_model(pretrained_file,model1)
#out1=model1(input_tensor)
#print("out1:",out1.shape,out1[0,0:10])
#
new_file="new_model.pth"
model=resnet18()
new_model=modify_model(pretrained_file,model,old_prefix=["layer4"],new_prefix=["layer44"])
torch.save(new_model.state_dict(),new_file)
model2=resnet18()
model2.load_state_dict(torch.load(new_file))
model2.eval()
out2=model2(input_tensor)
print("out2:",out2.shape,out2[0,0:10])
这时，输出，跟之前一模一样了。
out:torch.Size([1,1000])tensor([0.4010,0.8436,0.3072,0.0627,0.4446,0.8470,0.1882,0.7012,0.2988,-0.7574],grad_fn=
)
3.去除原模型的某些模块
下面是在不修改原模型代码的情况下,通过"resnet18.named_children()"和"resnet18.children()"的方法去除子模块"fc"和"avgpool"
importtorch
importtorchvision.modelsasmodels
fromcollectionsimportOrderedDict
if__name__=="__main__":
resnet18=models.resnet18(False)
print("resnet18",resnet18)
#usenamed_children()
resnet18_v1=OrderedDict(resnet18.named_children())
#removeavgpool,fc
resnet18_v1.pop("avgpool")
resnet18_v1.pop("fc")
resnet18_v1=torch.nn.Sequential(resnet18_v1)
print("resnet18_v1",resnet18_v1)
#usechildren
resnet18_v2=torch.nn.Sequential(*list(resnet18.children())[:-2])
print(resnet18_v2,resnet18_v2)
补充：pytorch导入(部分)模型参数
背景介绍：
我的想法是把一个预训练的网络的参数导入到我的模型中，但是预训练模型的参数只是我模型参数的一小部分，怎样导进去不出差错了，请来听我说说。
解法
首先把你需要添加参数的那一小部分模型提取出来，并新建一个类进行重新定义，如图向Alexnet中添加前三层的参数，重新定义前三层。
接下来就是导入参数
checkpoint=torch.load(config.pretrained_model)
#changenameandloadparameters
model_dict=model.net1.state_dict()
checkpoint={k.replace('features.features','featureExtract1'):vfork,vincheckpoint.items()}
checkpoint={k:vfork,vincheckpoint.items()ifkinmodel_dict.keys()}
model_dict.update(checkpoint)
model.net1.load_state_dict(model_dict)
程序如上图所示，主要是第三、四句，第三是替换，别人训练的模型参数的键和自己的定义的会不一样，所以需要替换成自己的；第四句有个if用于判断导入需要的参数。其他语句都相当于是模板，套用即可。
以上为个人经验，希望能给大家一个参考，也希望大家多多支持毛票票。如有错误或未考虑完全的地方，望不吝赐教。
声明：本文内容来源于网络，版权归原作者所有，内容由互联网用户自发贡献自行上传，本网站不拥有所有权，未作人工编辑处理，也不承担相关法律责任。如果您发现有涉嫌版权的内容，欢迎发送邮件至：czq8825#qq.com（发邮件时，请将#更换为@）进行举报，并提供相关证据，一经查实，本站将立刻删除涉嫌侵权内容。