深度学习中简易FC和CNN搭建

• TensorFlow是由谷歌开发的
• PyTorch是由Facebook人工智能研究院（Facebook AI Research）开发的

Torch和cuda版本的对应，手动安装较好

全连接FC(Batch*Num)

搭建建议网络：

from torch import nn

class Mnist_NN(nn.Module):
    def __init__(self):
        super().__init__()
        self.hidden1 = nn.Linear(784, 128)
        self.hidden2 = nn.Linear(128, 256)
        self.out  = nn.Linear(256, 10)

    def forward(self, x):
        x = F.relu(self.hidden1(x))
        x = F.relu(self.hidden2(x))
        x = self.out(x)
        return x

封装数据

from torch.utils.data import TensorDataset
from torch.utils.data import DataLoader

train_ds = TensorDataset(x_train, y_train)
train_dl = DataLoader(train_ds, batch_size=bs, shuffle=True)

valid_ds = TensorDataset(x_valid, y_valid)
valid_dl = DataLoader(valid_ds, batch_size=bs * 2)

def get_data(train_ds, valid_ds, bs):
    return (
        DataLoader(train_ds, batch_size=bs, shuffle=True),
        DataLoader(valid_ds, batch_size=bs * 2),
    )

训练模型：

import numpy as np

def fit(steps, model, loss_func, opt, train_dl, valid_dl):
    for step in range(steps):
        model.train()
        for xb, yb in train_dl:
            loss_batch(model, loss_func, xb, yb, opt)

        model.eval()
        with torch.no_grad():
            losses, nums = zip(
                *[loss_batch(model, loss_func, xb, yb) for xb, yb in valid_dl]
            )
        val_loss = np.sum(np.multiply(losses, nums)) / np.sum(nums)
        print('当前step:'+str(step), '验证集损失：'+str(val_loss))

一般在训练模型时加上model.train()，这样会正常使用Batch Normalization和 Dropout；测试的时候一般选择model.eval()，这样就不会使用Batch Normalization和 Dropout

批量损失函数

from torch import optim
def get_model():
    model = Mnist_NN()
    return model, optim.SGD(model.parameters(), lr=0.001)

def loss_batch(model, loss_func, xb, yb, opt=None):
    loss = loss_func(model(xb), yb)

    if opt is not None:
        loss.backward()
        opt.step()
        opt.zero_grad()

    return loss.item(), len(xb)

优化器
SGD是一种简单且易于实现的优化算法，但在大规模数据集和复杂模型上收敛缓慢。
Adam是一种自适应学习率调整的优化算法，能够更快地收敛，但可能会占用更多的内存。
在实践中，根据具体问题和数据集的特点，选择适合的优化算法可以提高训练效果。

卷积神经网络CNN(Batch * C * H * W)

Channel First
引入py库

import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
from torchvision import datasets,transforms 
import matplotlib.pyplot as plt
import numpy as np

预处理

# 定义超参数 
input_size = 28  #图像的总尺寸28*28
num_classes = 10  #标签的种类数
num_epochs = 3  #训练的总循环周期
batch_size = 64  #一个撮（批次）的大小，64张图片

# 训练集
train_dataset = datasets.MNIST(root='./data',  
                            train=True,   
                            transform=transforms.ToTensor(),  
                            download=True) 

# 测试集
test_dataset = datasets.MNIST(root='./data', 
                           train=False, 
                           transform=transforms.ToTensor())

# 构建batch数据
train_loader = torch.utils.data.DataLoader(dataset=train_dataset, 
                                           batch_size=batch_size, 
                                           shuffle=True)
test_loader = torch.utils.data.DataLoader(dataset=test_dataset, 
                                           batch_size=batch_size, 
                                           shuffle=True)

构建CNN

class CNN(nn.Module):
    def __init__(self):
        super(CNN, self).__init__()
        self.conv1 = nn.Sequential(         # 输入大小 (1, 28, 28)
            nn.Conv2d(
                in_channels=1,              # 灰度图
                out_channels=16,            # 要得到几多少个特征图
                kernel_size=5,              # 卷积核大小
                stride=1,                   # 步长
                padding=2,                  # 如果希望卷积后大小跟原来一样，需要设置padding=(kernel_size-1)/2 if stride=1
            ),                              # 输出的特征图为 (16, 28, 28)
            nn.ReLU(),                      # relu层
            nn.MaxPool2d(kernel_size=2),    # 进行池化操作（2x2 区域）, 输出结果为： (16, 14, 14)
        )
        self.conv2 = nn.Sequential(         # 下一个套餐的输入 (16, 14, 14)
            nn.Conv2d(16, 32, 5, 1, 2),     # 输出 (32, 14, 14)
            nn.ReLU(),                      # relu层
            nn.Conv2d(32, 32, 5, 1, 2),
            nn.ReLU(),
            nn.MaxPool2d(2),                # 输出 (32, 7, 7)
        )
        
        self.conv3 = nn.Sequential(         # 下一个套餐的输入 (16, 14, 14)
            nn.Conv2d(32, 64, 5, 1, 2),     # 输出 (32, 14, 14)
            nn.ReLU(),             # 输出 (32, 7, 7)
        )
        self.out = nn.Linear(64 * 7 * 7, 10)   # 全连接层得到的结果

    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        x = self.conv3(x)
        x = x.view(x.size(0), -1)           # flatten操作，结果为：(batch_size, 32 * 7 * 7)
        output = self.out(x)
        return output

定义准确率

def accuracy(predictions, labels):
    pred = torch.max(predictions.data, 1)[1] 
    rights = pred.eq(labels.data.view_as(pred)).sum() 
    return rights, len(labels) 

训练网络模型

# 实例化
net = CNN() 
#损失函数
criterion = nn.CrossEntropyLoss() 
#优化器
optimizer = optim.Adam(net.parameters(), lr=0.001) #定义优化器，普通的随机梯度下降算法

#开始训练循环
for epoch in range(num_epochs):
    #当前epoch的结果保存下来
    train_rights = [] 
    
    for batch_idx, (data, target) in enumerate(train_loader):  #针对容器中的每一个批进行循环
        net.train()                             
        output = net(data) 
        loss = criterion(output, target) 
        optimizer.zero_grad() 
        loss.backward() 
        optimizer.step() 
        right = accuracy(output, target) 
        train_rights.append(right) 

    
        if batch_idx % 100 == 0: 
            
            net.eval() 
            val_rights = [] 
            
            for (data, target) in test_loader:
                output = net(data) 
                right = accuracy(output, target) 
                val_rights.append(right)
                
            #准确率计算
            train_r = (sum([tup[0] for tup in train_rights]), sum([tup[1] for tup in train_rights]))
            val_r = (sum([tup[0] for tup in val_rights]), sum([tup[1] for tup in val_rights]))

            print('当前epoch: {} [{}/{} ({:.0f}%)]\t损失: {:.6f}\t训练集准确率: {:.2f}%\t测试集正确率: {:.2f}%'.format(
                epoch, batch_idx * batch_size, len(train_loader.dataset),
                100. * batch_idx / len(train_loader), 
                loss.data, 
                100. * train_r[0].numpy() / train_r[1], 
                100. * val_r[0].numpy() / val_r[1]))