PyTorch深度学习快速入门（b站小土堆）P16-笔记
PyTorch深度学习快速入门（b站小土堆）P8-15笔记
torch.nn
import torch
from torch import nn
class TUdui(nn.Module): #继承Module
def __init__(self):
super().__init__()
def forward(self,input):
output = input + 1
return output
#实例化
tudui = Tudui()
x = torch.tensor(1.0)
output = tudui(x)
print(output)卷积操作
Stride=1,其实是控制了卷积核横向和纵向的移动都是1步
import torch
import torch.nn.functional as F
#输入
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]])
#卷积核
kernel = torch.tensor([[1,2,1],
[0,1,0],
[2,1,0]])
#使用reshape
input = torch.reshape(input,(1,1,5,5))
kernel = torch.reshape(kernel,(1,1,3,3))
print(input.shape)
print(kernel.shape)
#输出
output1 = F.conv2d(input,kernel,stride=1)
print(output1)
output2 = F.conv2d(input,kernel,stride=2)
print(output2)
output3 = F.conv2d(input,kernel,stride=1,padding=1)
print(output3)
padding
padding = 1 时，如下↓，一般空白处是填充数字0
卷积层
import torch
import torchvision
from torch import nn
from torch.nn import Conv2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
#测试集
dataset = torchvision.datasets.CIFAR10("../data",train=False,transform=torchvision.transforms.ToTensor(),
download=True)
#dataloader
dataloader = DataLoader(dataset,batch_size=64)
#卷积操作
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
writer = SummaryWriter("logs")
#实例化
tudui = Tudui()
step = 0
for data in dataloader:
imgs,tergets = data
output = tudui(imgs)
print(imgs.shape) #torch.Size([64,3,32,32])
print(output.shape) #torch.Size([64,6,30,30])
writer.add_image("input",imgs,step,dataformats='NCHW')
#因为output是六通道，所以要变成3通道才能显示，(XX,3,30,30)不知道XX填什么就写-1，它会自动计算的。
output = torch.reshape(output,(-1,3,30,30))
writer.add_image("output", output, step,dataformats='NCHW')
step=step+1
最大池化的使用
步数默认与池化核相同，如下图中的 Stride = 3
Ceil_model如果为True，则表示保留
import torch
from torch import nn
from torch.nn import MaxPool2d
#输入
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)
#dtype=torch.float32,就是把数据变成浮点数，否则会报错
input = torch.reshape(input,(-1,1,5,5))
#要是用reshape是因为池化操作的input是要四维的（N,C,H,W）
#池化
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.maxpool1 = MaxPool2d(kernel_size=3,ceil_mode=True)#核的大小3x3
def forward(self,input):
output = self.maxpool1(input)
return output
#实例化
tudui = Tudui()
output = tudui(input)
print(output)
#输出tensor([[[[2., 3.],
# [5., 1.]]]])最大池化（使用数据集CIFAR10）
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)
#池化
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.maxpool1 = MaxPool2d(kernel_size=3,ceil_mode=True)#核的大小3x3
def forward(self,input):
output = self.maxpool1(input)
return output
writer = SummaryWriter("logs_maxpool")
#实例化
tudui = Tudui()
step = 0
for data in dataloader:
imgs,targets = data
writer.add_image("input", imgs, step)
output = tudui(imgs)
writer.add_image("output",output,step)
step=step+1
writer.close()
非线性激活
ReLU
inplace=True 或者 False
代码
import torch
from torch import nn
from torch.nn import ReLU
#输入
input = torch.tensor([[1,-0.5],
[-1,3]])
input = torch.reshape(input,(-1,1,2,2))
#ReLU
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.relu1 = ReLU()
def forward(self,input):
output = self.relu1(input)
return output
#实例化
tudui = Tudui()
output = tudui(input)
print(output)sigmoid
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
#输入
dataset = torchvision.datasets.CIFAR10("data",train=False,transform=torchvision.transforms.ToTensor(),
download=True)
dataloader = DataLoader(dataset,batch_size=64)
#Sigmoid
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.sigmoid = Sigmoid()
def forward(self,input):
output = self.sigmoid(input)
return output
writer = SummaryWriter("logs_s")
step=0
#实例化
tudui = Tudui()
for data in dataloader:
imgs,targets = data
writer.add_images("input",imgs,step)
output = tudui(imgs)
writer.add_images("output",output,step)
step=step+1
writer.close()
线性层
代码
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.linear = Linear(196608,10)
def forward(self,input):
output = self.linear(input)
return output
tudui = Tudui()
for data in dataloader:
imgs,targets = data
print(imgs.shape) #原来的结果：[64,3,32,32]
output1 = torch.flatten(imgs)
print(output1.shape) #经过flatten后的，结果：[196608]
output2 = tudui(output1)
print(output2.shape) #经过linear的结果：[10]
搭建小实战和Sequential的使用
padding是被计算出来的，padding=2
代码
import torch
from torch import nn
from torch.nn import Conv2d, MaxPool2d, Flatten, Linear
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.conv1 = Conv2d(3, 32, 5, padding=2) # padding是算出来的
self.maxpool1 = MaxPool2d(2)
self.conv2 = Conv2d(32, 32, 5, padding=2) # padding是算出来的
self.maxpool2 = MaxPool2d(2)
self.conv3 = Conv2d(32, 64, 5, padding=2) # padding是算出来的
self.maxpool3 = MaxPool2d(2)
self.flatten = Flatten()
self.linear1 = Linear(1024, 64)
self.linear2 = Linear(64, 10)
def forward(self, x):
x = self.conv1(x)
x = self.maxpool1(x)
x = self.conv2(x)
x = self.maxpool2(x)
x = self.conv3(x)
x = self.maxpool3(x)
x = self.flatten(x)
x = self.linear1(x)
x = self.linear2(x)
return x
tudui = Tudui()
#验证
input = torch.ones((64,3,32,32))
output = tudui(input)
print(output.shape) #torch.Size([64, 10])使用Sequential
mport torch
from torch import nn
from torch.nn import Conv2d, MaxPool2d, Flatten, Linear, Sequential
from torch.utils.tensorboard import SummaryWriter
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.model1 = Sequential(
Conv2d(3, 32, 5, padding=2),
MaxPool2d(2),
Conv2d(32, 32, 5, padding=2),
MaxPool2d(2),
Conv2d(32, 64, 5, padding=2),
MaxPool2d(2),
Flatten(),
Linear(1024, 64),
Linear(64, 10)
)
def forward(self, x):
x = self.model1(x)
return x
tudui = Tudui()
#验证
input = torch.ones((64,3,32,32))
output = tudui(input)
print(output.shape) #torch.Size([64, 10])
writer = SummaryWriter("log_seq")
writer.add_graph(tudui,input)
writer.close()
损失函数与反向传播
代码
import torch
from torch import nn
from torch.nn import L1Loss
inputs = torch.tensor([1,2,3],dtype=torch.float32) #要是浮点型
targets = torch.tensor([1,2,5],dtype=torch.float32)
#reshape
inputs = torch.reshape(inputs,(1,1,1,3))
targets = torch.reshape(targets,(1,1,1,3))
#损失
#L1Loss
loss = L1Loss()
result = loss(inputs,targets)
print(result)
#MSELoss
loss_mse = nn.MSELoss()
result_mes = loss_mes(inputs,targets)
print(result_mes)CrossEntropyLoss
#CrossEntropyLoss
x = torch.tensor([0.1,0.2,0.3])
y = torch.tensor([1])
x = torch.reshape(x,(1,3)) #batch_size=1 , 共有3个分类
loss_cross = nn.CrossEntropyLoss()
result_cross = loss_cross(x,y)
print(result_cross)
优化器
import torchvision
import torch
from torch import nn
from torch.nn import Sequential, Conv2d, MaxPool2d, Flatten,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=1)
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.model = Sequential(
Conv2d(3, 32, 5, padding=2),
MaxPool2d(2),
Conv2d(32, 32, 5, padding=2),
MaxPool2d(2),
Conv2d(32, 64, 5, padding=2),
MaxPool2d(2),
Flatten(),
Linear(1024,64),
Linear(64,10))
def forward(self,x):
x = self.model(x)
return x
tudui = Tudui()
#定义loss
loss = nn.CrossEntropyLoss()
#优化器
optim = torch.optim.SGD(tudui.parameters(),lr=0.01)#参数、学习速率
for epoch in range(20):#20轮循环
running_loss = 0.0
for data in dataloader:
imgs,targets = data
output = tudui(imgs)
result_loss = loss(output,targets)
optim.zero_grad() #梯度清零
result_loss.backward() #反向传播
optim.step()#对每个参数进行调优
running_loss = running_loss + result_loss
print(result_loss)
现有网络模型的使用及修改
import torchvision
from torch import nn
vgg16_false = torchvision.models.vgg16(pretrained=False)
vgg16_true = torchvision.models.vgg16(pretrained=True)
#false 表示加载网络模型
#true 表示有下载参数
print(vgg16_true)
#在现有的网络中加模型
vgg16_true.add_module("add_linear",nn.Linear(1000,10))
vgg16_true.classifier.add_module("add_linear",nn.Linear(1000,10))
print(vgg16_true)
#修改模型
print(vgg16_false)
vgg16_false.classifier[6] = nn.Linear(4096,10)
print(vgg16_false)
网络模型的保存与读取
1.保存
import torch
import torchvision
from torch import nn
vgg16 = torchvision.models.vgg16(pretrained=False)
#保存方式一，模型结构+模型参数
torch.save(vgg16,"vgg16_method1.pth")
#保存方式二，模型参数（官方推荐），使用的空间更小
torch.save(vgg16.state_dict(),"vgg16_method2.pth")
#保存自己定义的文件
class Tudui(nn.Module):
def __init__(self):
super(Tudui, self).__init__()
self.conv1 = nn.Conv2d(3,64,kernel_size=3)
def forward(self,x):
x = self.conv1(x)
return x
tudui = Tudui()
torch.save(tudui,"tudui_method.pth")2.加载
import torch
import torchvision
from model_save import *
#加载模型方式一
model = torch.load("vgg16_method1.pth")
print(model)
#加载模型方式二
vgg16 = torchvision.models.vgg16(pretrained=False)
vgg16.load_state_dict(torch.load("vgg16_method2.pth"))
print(vgg16)
#如果模型是自己定义的，怎么加载？
#在最前面引入 from model_save import * (model_save是我们保存模型的py文件)
model_tudui = torch.add("tudui_method.pth")
print(model_tudui)