布尔网络的稳定性的定义Day158:网络的参数

导读importtorch.nnasnnfromcollectionsimportOrderedDictclassModel(nn.Module):def__i....

import torch.nn as nnfrom collections import OrderedDictclass Model(nn.Module): def __init__(self): super(Model, self).__init__() self.conv1 = nn.Conv2d(in_channels=3, out_channels=64, kernel_size=3) self.conv2 = nn.Conv2d(64, 64, 3) self.maxpool1 = nn.MaxPool2d(2, 2) self.features = nn.Sequential(OrderedDict([ (conv3, nn.Conv2d(64, 128, 3)), (conv4, nn.Conv2d(128, 128, 3)), (relu1, nn.ReLU()) ])) def forward(self, x): x = self.conv1(x) x = self.conv2(x) x = self.maxpool1(x) x = self.features(x) return xm = Model()for p in m.parameters(): print(type(p.data), p.size())，今天小编就来说说关于布尔网络的稳定性的定义?下面更多详细答案一起来看看吧!

布尔网络的稳定性的定义

parameters方法

• 方法parameters()返回包含模型所有参数的迭代器，包含网络中的所有权值矩阵参数以及偏置参数。

import torch.nn as nnfrom collections import OrderedDictclass Model(nn.Module): def __init__(self): super(Model, self).__init__() self.conv1 = nn.Conv2d(in_channels=3, out_channels=64, kernel_size=3) self.conv2 = nn.Conv2d(64, 64, 3) self.maxpool1 = nn.MaxPool2d(2, 2) self.features = nn.Sequential(OrderedDict([ (conv3, nn.Conv2d(64, 128, 3)), (conv4, nn.Conv2d(128, 128, 3)), (relu1, nn.ReLU()) ])) def forward(self, x): x = self.conv1(x) x = self.conv2(x) x = self.maxpool1(x) x = self.features(x) return xm = Model()for p in m.parameters(): print(type(p.data), p.size())

其输出为：

<class torch.Tensor> torch.Size([64, 3, 3, 3])<class torch.Tensor> torch.Size([64])<class torch.Tensor> torch.Size([64, 64, 3, 3])<class torch.Tensor> torch.Size([64])<class torch.Tensor> torch.Size([128, 64, 3, 3])<class torch.Tensor> torch.Size([128])<class torch.Tensor> torch.Size([128, 128, 3, 3])<class torch.Tensor> torch.Size([128])

• 一般用来当作optimizer的参数，因为对网络进行训练时，需要将parameters()作为优化器optimizer的参数。

optimizer = torch.optim.SGD(m1.parameters(), lr = args.lr, momentum=args.momentum, weight_decay=args.weight_decay)

• parameters()返回网络的所有参数，主要是提供给optimizer用的。而要取得网络某一层的参数或者对参数进行一些特殊的处理（如fine-tuning)，则使用named_parameters()更为方便些，named_parameters()返回参数的名称及参数本身，可以按照参数名对一些参数进行处理。

m = Model()for k, v in m.named_parameters(): print(k, v.size())

输出为：

conv1.weight - torch.Size([64, 3, 3, 3])conv1.bias - torch.Size([64])conv2.weight - torch.Size([64, 64, 3, 3])conv2.bias - torch.Size([64])features.conv3.weight - torch.Size([128, 64, 3, 3])features.conv3.bias - torch.Size([128])features.conv4.weight - torch.Size([128, 128, 3, 3])features.conv4.bias - torch.Size([128])

再以上面的vgg网络为例：

for k,v in m1.named_parameters(): print(k,v.size())

named_parameters返回的是键值对，k为参数的名称 ，v为参数本身。输出结果为：

vgg.0.weight torch.Size([64, 3, 3, 3])vgg.0.bias torch.Size([64])vgg.2.weight torch.Size([64, 64, 3, 3])vgg.2.bias torch.Size([64])vgg.5.weight torch.Size([128, 64, 3, 3])vgg.5.bias torch.Size([128])vgg.7.weight torch.Size([128, 128, 3, 3])vgg.7.bias torch.Size([128])vgg.10.weight torch.Size([256, 128, 3, 3])vgg.10.bias torch.Size([256])vgg.12.weight torch.Size([256, 256, 3, 3])vgg.12.bias torch.Size([256])vgg.14.weight torch.Size([256, 256, 3, 3])vgg.14.bias torch.Size([256])vgg.17.weight torch.Size([512, 256, 3, 3])vgg.17.bias torch.Size([512])vgg.19.weight torch.Size([512, 512, 3, 3])vgg.19.bias torch.Size([512])vgg.21.weight torch.Size([512, 512, 3, 3])vgg.21.bias torch.Size([512])vgg.24.weight torch.Size([512, 512, 3, 3])vgg.24.bias torch.Size([512])vgg.26.weight torch.Size([512, 512, 3, 3])vgg.26.bias torch.Size([512])vgg.28.weight torch.Size([512, 512, 3, 3])vgg.28.bias torch.Size([512])

参数名的命名规则：属性名称.参数所属层的编号.weight/bias。 在fine-tuning的时候，给一些特定层的参数赋值是非常方便的，这点在后面在加载预训练模型时会看到。