神經網絡 非線性激活層

作用：增強模型的非線性擬合能力

非線性激活層網絡：

class activateNet(nn.Module):def __init__(self):super(activateNet,self).__init__()self.relu = nn.ReLU()self.sigmoid = nn.Sigmoid()def forward(self,input):#output = self.relu(input)output = self.sigmoid(input)return outputactivatenet = activateNet()

relu：

sigmoid：

完整代碼：

這里是將上一步池化層的輸出作為非線性激活層的輸入

code：

import torch
import torchvision
from torch import nn
from torch.nn import Conv2d, MaxPool2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
from torchvision.datasets import ImageFolder
from torchvision import transforms#數據預處理
transform = transforms.Compose([transforms.Resize((224,224)),transforms.ToTensor(),transforms.Normalize(mean = [0.5,0.5,0.5],std = [0.5,0.5,0.5])
])#加載數據集
folder_path = '../images'
dataset = ImageFolder(folder_path,transform=transform)
dataloader = DataLoader(dataset,batch_size=1)#卷積
class convNet(nn.Module):def __init__(self):#調用父類nn.Module的構造函數super(convNet,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 xconvnet = convNet()#池化
class poolNet(nn.Module):def __init__(self):super(poolNet,self).__init__()#ceil_mode=True表示邊緣不滿3x3的部分也會被池化#kernel_size=3 默認是卷積核的大小self.maxpool1 = MaxPool2d(kernel_size=3,ceil_mode=True)self.maxpool2 = MaxPool2d(kernel_size=3,ceil_mode=False)def forward(self,input):output = self.maxpool1(input)#output = self.maxpool2(input)return outputpoolnet = poolNet()#非線性激活層（這里使用relu、sigmoid）
class activateNet(nn.Module):def __init__(self):super(activateNet,self).__init__()self.relu = nn.ReLU()self.sigmoid = nn.Sigmoid()def forward(self,input):#output = self.relu(input)output = self.sigmoid(input)return outputactivatenet = activateNet()writer = SummaryWriter('../logs')cnt = 0
for data in dataloader:img,label = dataprint(img.shape)conv_output = convnet(img)print(conv_output.shape)writer.add_images('input',img,cnt)conv_output = torch.reshape(conv_output,(-1,3,222,222))writer.add_images('conv_output',conv_output,cnt)pool_output = poolnet(conv_output)writer.add_images('pool_output',pool_output,cnt)activate_output = activatenet(pool_output)writer.add_images('activate_output',activate_output,cnt)cnt = cnt + 1writer.close()

神經網絡 正則化層

防止過擬合

比較簡單 不寫了

這這里區別幾個概念：

- 歸一化：將數據映射到特定區間（如[0,1]或[-1,1]）的過程，常見方法如min-max歸一化，通過減去最小值并除以極差實現。其目的是消除不同特征量綱差異的影響，使數據處于統一量級，便于模型快速收斂，例如將圖像像素值從0-255轉換為0-1。

- 標準化：將數據轉換為均值為0、標準差為1的分布，即通過減去均值并除以標準差實現（z-score標準化）。它能讓數據更符合正態分布，降低異常值對模型的影響，常用于線性回歸、SVM等對數據分布敏感的算法中。

- 正則化：在模型訓練中加入額外約束（如L1、L2范數）以防止過擬合的技術。L1正則化通過懲罰權重絕對值，可產生稀疏權重；L2正則化（權重衰減）懲罰權重平方，使權重更平滑。其核心是平衡模型復雜度與擬合能力，提升泛化性能。

神經網絡 線性層（全連接層）

輸入為上一步非線性激活層的輸出

線性層：

code:

class linearNet(nn.Module):def __init__(self):super(linearNet,self).__init__()#in_features(根據下面展開后輸出的數值) 表示輸入特征的維度，out_features=10 表示輸出特征的維度self.linear = nn.Linear(32856,10)def forward(self,input):output = self.linear(input)return outputlinearnet = linearNet()

in_features:

    #為了得到in_featuresprint(torch.reshape(activate_output,(1,1,1,-1)).shape)

完整代碼：

code:

import torch
import torchvision
from torch import nn
from torch.nn import Conv2d, MaxPool2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
from torchvision.datasets import ImageFolder
from torchvision import transforms#數據預處理
transform = transforms.Compose([transforms.Resize((224,224)),transforms.ToTensor(),transforms.Normalize(mean = [0.5,0.5,0.5],std = [0.5,0.5,0.5])
])#加載數據集
folder_path = '../images'
dataset = ImageFolder(folder_path,transform=transform)
#drop_last=True 表示最后一個batch可能小于batch_size，不足的部分會被丟棄
#dataloader = DataLoader(dataset,batch_size=1,drop_last=True)
dataloader = DataLoader(dataset,batch_size=1)#卷積
class convNet(nn.Module):def __init__(self):#調用父類nn.Module的構造函數super(convNet,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 xconvnet = convNet()#池化
class poolNet(nn.Module):def __init__(self):super(poolNet,self).__init__()#ceil_mode=True表示邊緣不滿3x3的部分也會被池化#kernel_size=3 默認是卷積核的大小self.maxpool1 = MaxPool2d(kernel_size=3,ceil_mode=True)self.maxpool2 = MaxPool2d(kernel_size=3,ceil_mode=False)def forward(self,input):output = self.maxpool1(input)#output = self.maxpool2(input)return outputpoolnet = poolNet()#非線性激活層（這里使用relu、sigmoid）
class activateNet(nn.Module):def __init__(self):super(activateNet,self).__init__()self.relu = nn.ReLU()self.sigmoid = nn.Sigmoid()def forward(self,input):#output = self.relu(input)output = self.sigmoid(input)return outputactivatenet = activateNet()#線性層（全連接層）
class linearNet(nn.Module):def __init__(self):super(linearNet,self).__init__()#in_features(根據下面展開后輸出的數值) 表示輸入特征的維度，out_features=10 表示輸出特征的維度self.linear = nn.Linear(32856,10)def forward(self,input):output = self.linear(input)return outputlinearnet = linearNet()writer = SummaryWriter('../logs')cnt = 0
for data in dataloader:img,label = dataprint(img.shape)conv_output = convnet(img)print(conv_output.shape)writer.add_images('input',img,cnt)conv_output = torch.reshape(conv_output,(-1,3,222,222))writer.add_images('conv_output',conv_output,cnt)pool_output = poolnet(conv_output)writer.add_images('pool_output',pool_output,cnt)activate_output = activatenet(pool_output)writer.add_images('activate_output',activate_output,cnt)#為了得到in_featuresprint(torch.reshape(activate_output,(1,1,1,-1)).shape)#flatten:展開成一維數組linear_output = torch.flatten(activate_output)linear_output = linearnet(linear_output)print(linear_output.shape)cnt = cnt + 1writer.close()