【土堆pytorch实战】P17-P21卷积池化线性层
P17 nn.conv
1. 卷积基础
Conv2d 二维卷积
官方文档：https://pytorch.org/docs/1.9.0/generated/torch.nn.functional.conv2d.html#torch.nn.functional.conv2d
- CLASS torch.nn.Conv2d(in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True, padding_mode=‘zeros’, device=None, dtype=None)
stride表示步长，padding表示边缘填充
import torch
import torch.nn.functional as F
input=torch.tensor([[1,2,0,3,1],
[0,1,2,3,1],
[1,2,1,0,0],
[5,2,3,1,1],
[2,1,0,1,1]])
kernel=torch.tensor([[1,2,1],
[0,1,0],
[2,1,0]])
print(input.shape) #shape为 torch.size([5,5]) 两个参数
print(kernel.shape) #shape为 torch.size([3,3]) 两个参数
# 由于官方文档中要求input是四个参数（minibatch,channel,H,W) 故使用reshape 两参变四参
input=torch.reshape(input,(1,1,5,5))
kernel=torch.reshape(kernel,(1,1,3,3))
output=F.conv2d(input,kernel,stride=1) #步长为1
#卷积后的结果
print(output)
2. 在数据集上使用卷积操作
import torchvision
from torch import nn
from torch.nn import Conv2d
from torch.utils.data import DataLoader
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.conv1=Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)
def forward(self,x):
x=self.conv1(x)
return x
tudui=Tudui()
print(tudui)
for data in dataloader:
imgs,targets=data
output=tudui(imgs)
print(imgs.shape)
print(output.shape)
输入图像从[64,3,32,32] 经过卷积操作变成了[64,6,30,30]
- 使用tensorboard查看
import torchvision
from torch import nn
from torch.nn import Conv2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
import torch
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.conv1=Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)
def forward(self,x):
x=self.conv1(x)
return x
tudui=Tudui()
print(tudui)
writer=SummaryWriter("conv2d")
step=0
for data in dataloader:
imgs,targets=data
output=tudui(imgs)
writer.add_images("input",imgs,step)
#torch.size([64,6,30,30]) -->[xxx,3,30,30] 6个通道变成3个通道
output=torch.reshape(output,(-1,3,30,30)) #batchsize=-1,自动计算
writer.add_images("outout",output,step)
step=step+1
P19 池化层
maxpool 也叫下采样
maxunpool 也叫上采样
- CLASS torch.nn.MaxPool2d(kernel_size, stride=None, padding=0, dilation=1, return_indices=False, ceil_mode=False)
ceil_mode: ceil 天花板 floor 地板 ceil_mode=True, 边缘也取
stride: 默认kernel_size
最大池化作用：保留数据特征，减少数据量
1.最大池化基础
import torch
from torch import nn
from torch.nn import MaxPool2d
input=torch.tensor([[1,2,0,3,1],
[0,1,2,3,1,],
[1,2,1,0,0],
[5,2,3,1,1],
[2,1,0,1,1]],dtype=torch.float32)
input=torch.reshape(input,(-1,1,5,5))
print(input.shape)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.maxpool1=MaxPool2d(kernel_size=3,ceil_mode=True)
def forward(self,input):
output=self.maxpool1(input)
return output
tudui=Tudui()
output=tudui(input)
print(output)
2.在数据集上使用最大池化操作
import torch
import torchvision
from torch import nn
from torch.nn import MaxPool2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.maxpool1=MaxPool2d(kernel_size=3,ceil_mode=True)
def forward(self,input):
output=self.maxpool1(input)
return output
tudui=Tudui()
writer=SummaryWriter("log_maxpool")
step=0
for data in dataloader:
imgs,targets=data
writer.add_images("input_maxpool",imgs,step)
output=tudui(imgs)
writer.add_images("output_maxpool",output,step)
step=step+1
writer.close()
P20 非线性激活层
非线性激活的目的是激活模型，提高模型的泛化能力
https://pytorch.org/docs/1.9.0/nn.html#non-linear-activations-weighted-sum-nonlinearity
- relu
import torch
from torch import nn
from torch.nn import ReLU
input=torch.tensor([[1,-0.5],
[-1,3]])
output=torch.reshape(input,(-1,1,2,2))
print(output.shape)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.relu1=ReLU() #inplace是否进行原地操作 默认false 可以保留原数据
def forward(self,input):
output=self.relu1(input)
return output
tudui=Tudui()
output=tudui(input)
print(output)
- Sigmoid
在tensorboard中显示
import torch
import torchvision
from torch import nn
from torch.nn import ReLU, Sigmoid
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.sigmoid1=Sigmoid()
def forward(self,input):
output=self.sigmoid1(input)
return output
tudui=Tudui()
writer=SummaryWriter("logs_activate")
step=0
for data in dataloader:
imgs,targets=data
writer.add_images("input_act",imgs,step)
output=tudui(imgs)
writer.add_images("output_act",output,step)
step+=1
writer.close()
P21 线性层及其他层
正则化层：加快神经网络训练速度，同时防止模型过于复杂，防止过拟合
recurrent layerRNN：用于 文字识别（其中LSTM用于nlp）
dropout layers: 随机失活，防止过拟合
- linear layers
import torch
import torchvision
from torch.utils.data import DataLoader
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)
for data in dataloader:
imgs,targets=data
print(imgs.shape)
output=torch.reshape(imgs,(1,1,1,-1))
print(output.shape)
原始图片的size(64,3,32,32)，reshape后size变为（1，1，1，196608）
经过线性层后变为（1，1，1，10）
import torch
import torchvision
from torch import nn
from torch.nn import Linear
from torch.utils.data import DataLoader
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)
class Tudui(nn.Module):
def __init__(self):
super(Tudui,self).__init__()
self.linear1=Linear(196608,10) #输入196608维 输出10维
def forward(self,input):
output=self.linear1(input)
return output
tudui=Tudui()
for data in dataloader:
imgs,targets=data
print(imgs.shape)
output=torch.reshape(imgs,(1,1,1,-1))
print(output.shape)
output=tudui(output)
print(output.shape)
使用flatten（）替代reshape
for data in dataloader:
imgs,targets=data
print(imgs.shape)
# output=torch.reshape(imgs,(1,1,1,-1))
output=torch.flatten(imgs)
print(output.shape)
output=tudui(output)
print(output.shape)
