基于pytorch的神经病网络搭建学习

1.pycharm中code方法的使用

1.1父类重写技巧

• 操作：在需要重写的方法上右键，选择code-->Generate > Override Methods。
• 作用：自动生成重写父类或接口的方法

2.简单神经网络

import torch
from  torch import  nn

class  yu(nn.Module):
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)
    def forward(self,input):
        output =input+1
        return output
hh =yu()
x =torch.tensor(1.0)
output =hh(x)
print(output)

3.卷积操作

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],
])
input =torch.reshape(input,(1,1,5,5))
kernel = torch.reshape(kernel,(1,1,3,3))
print(input.shape)
print(kernel.shape)
#这里进行reshape的目的是卷积要求输入四维（批次，通道，宽和高）
output = F.conv2d(input,kernel,stride=1,padding=1
)
#padding是填充的意思，stride是步长
print(output)

3.2卷积实战 

import torch
import torchvision
from requests.packages import target
from torch import nn
from torch.nn import Conv2d
from torch.onnx.symbolic_opset9 import conv2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

dataset_transform = torchvision.transforms.Compose({
    torchvision.transforms.ToTensor()
})
train_set = torchvision.datasets.CIFAR10(root="./dataset", train=False, transform=dataset_transform, download=False)

test_load =DataLoader(dataset=train_set,batch_size=64,shuffle=True,num_workers=0,drop_last=False)

class yu(nn.Module):
    def __init__(self, *args, **kwargs) -> None:
        super(yu,self).__init__(*args, **kwargs)
        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
hh=yu()
#print(hh)
writer = SummaryWriter("C:\yolo\yolov5-5.0\logs")
step =0
for data in test_load:
    imgs,targets = data
    output = hh(imgs)
    print(imgs.shape)
    print(output.shape)
    writer.add_images("input",imgs,step)
    output=torch.reshape(output,(-1,3,30,30))
    writer.add_images("output",output,step)
    step= step +1

4.池化操作：

池化：提取图片在每一个池化核中的最大值，得到输出一个经过提取的图像，保留一部分特征降低复杂度 

---dilation 空洞卷积

---ceil model的概念

import torch
from torch import nn
from torch.nn import MaxPool2d
from torch.nn.functional import DType

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))#批次大小，通道数，高度，宽度
print(input.shape)

class yu(nn.Module):

    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)
        self.maxpool1=MaxPool2d(kernel_size=3,ceil_mode=False)
    def forward(self,input):
        output = self.maxpool1(input)
        return  output
hh=yu()
output = hh(input)
print(output)

 5.非线性激活操作

import torch
from torch import nn
from torch.nn import MaxPool2d, ReLU
from torch.nn.functional import DType

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,-1,1,1]
    ]#,dtype=torch.float32
)
input=torch.reshape(input,(-1,1,5,5))#批次大小，通道数，高度，宽度
print(input.shape)





class yu(nn.Module):

    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)
        self.relul=ReLU()
    def forward(self,input):
        output = self.relul(input)
        return  output
hh=yu()
output = hh(input)
print(output)

谢谢~~~~~~