深度学习本科课程 实验2 前馈神经网络

任务 3.3 课程实验要求 （1）手动实现前馈神经网络解决上述回归、二分类、多分类任务
l 从训练时间、预测精度、Loss变化等角度分析实验结果（最好使用图表展示）
（2）利用torch.nn实现前馈神经网络解决上述回归、二分类、多分类任务 l 从训练时间、预测精度、Loss变化等角度分析实验结果（最好使用图表展示）
（3）在多分类任务中使用至少三种不同的激活函数 l 使用不同的激活函数，进行对比实验并分析实验结果
（4）对多分类任务中的模型评估隐藏层层数和隐藏单元个数对实验结果的影响
l 使用不同的隐藏层层数和隐藏单元个数，进行对比实验并分析实验结果

一、手动生成实现前馈神经网络解决回归、二分类、多分类任务

1.1 任务内容

分析实验结果并绘制训练集和测试集的loss曲线

1.2 任务思路及代码

# 导入模块
import torch
import torch.nn as nn
import numpy as np
import torchvision
from torchvision import transforms
import time

# 定义绘图函数
import matplotlib.pyplot as plt
def draw_loss(train_loss, test_loss):
    x = np.linspace(0, len(train_loss), len(train_loss))
    plt.plot(x, train_loss, label="Train Loss", linewidth=1.5)
    plt.plot(x, test_loss, label="Test Loss", linewidth=1.5)
    plt.xlabel("Epoch")
    plt.ylabel("Loss")
    plt.legend()
    plt.show()

# 定义评价函数
def evaluate_accuracy(data_iter, model, loss_func):
    acc_sum, test_l_sum, n, c = 0.0, 0.0, 0, 0
    for X, y in data_iter:
        result = model.forward(X)
        acc_sum += (result.argmax(dim=1)==y).float().sum().item()
        test_l_sum += loss_func(result, y).item()
        n += y.shape[0]
        c += 1
    return acc_sum/n, test_l_sum/c

# 回归任务
n_train = 7000
n_test = 3000
num_inputs = 500
true_w, true_b = torch.ones(num_inputs, 1)*0.0056, 0.028

# 生成数据集
features = torch.randn((n_train + n_test, num_inputs))
labels = torch.matmul(features, true_w) + true_b
labels += torch.tensor(np.random.normal(0, 0.01, size=labels.size()), dtype=torch.float)
# 数据划分
train_features, test_features = features[:n_train, :], features[n_train:, :]
train_labels, test_labels = labels[:n_train], labels[n_train:]
batch_size1 = 128

traindataset1 = torch.utils.data.TensorDataset(train_features, train_labels)
testdataset1 = torch.utils.data.TensorDataset(test_features, test_labels)

traindataloader1 = torch.utils.data.DataLoader(dataset=traindataset1, batch_size=batch_size1, shuffle=True)
testdataloader1 = torch.utils.data.DataLoader(dataset=testdataset1, batch_size=batch_size1, shuffle=False)

# 定义损失函数
def my_cross_entropy_loss(y_hat, labels):
    def log_softmax(y_hat):
        max_v = torch.max(y_hat, dim=1).values.unsqueeze(dim=1)
        return y_hat - max_v - torch.log(torch.exp(y_hat-max_v).sum(dim=1).unsqueeze(dim=1))

    return (-log_softmax(y_hat))[range(len(y_hat)), labels].mean()

# 定义优化算法
def SGD(params, lr):
    for param in params:
        param.data -= lr*param.grad

def mse(pred, true):
    ans = torch.sum((true-pred)**2)/len(pred)
    return ans

class Net1():
    def __init__(self):
        # 设置隐藏层和输出层的节点数
        num_inputs, num_hiddens, num_outputs = 500, 256, 1
        w_1 = torch.tensor(np.random.normal(0,0.01,(num_hiddens,num_inputs)),dtype=torch.float32,requires_grad=True)
        b_1 = torch.zeros(num_hiddens, dtype=torch.float32,requires_grad=True)
        w_2 = torch.tensor(np.random.normal(0, 0.01,(num_outputs, num_hiddens)),dtype=torch.float32,requires_grad=True)
        b_2 = torch.zeros(num_outputs,dtype=torch.float32, requires_grad=True)
        self.params = [w_1, b_1, w_2, b_2]

        # 定义模型结构
        self.input_layer = lambda x: x.view(x.shape[0],-1)
        self.hidden_layer = lambda x: self.my_relu(torch.matmul(x,w_1.t())+b_1)
        self.output_layer = lambda x: torch.matmul(x,w_2.t()) + b_2

    def my_relu(self, x):
        return torch.max(input=x,other=torch.tensor(0.0))

    def forward(self, x):
        flatten_input = self.input_layer(x)
        hidden_output = self.hidden_layer(flatten_input)
        final_output = self.output_layer(hidden_output)
        return final_output
        
# 训练
model1 = Net1()  # logistics模型
criterion = my_cross_entropy_loss
lr = 0.01   # 学习率
batchsize = 128 
epochs = 40 #训练轮数
train_all_loss1 = [] # 记录训练集上得loss变化
test_all_loss1 = [] #记录测试集上的loss变化
begintime1 = time.time()
for epoch in range(epochs):
    train_l = 0
    for data, labels in traindataloader1:
        pred = model1.forward(data)
        
        train_each_loss = mse(pred.view(-1,1), labels.view(-1,1)) #计算每次的损失值
        train_each_loss.backward() # 反向传播
        SGD(model1.params, lr) # 使用小批量随机梯度下降迭代模型参数
        # 梯度清零
        train_l += train_each_loss.item()
        for param in model1.params:
            param.grad.data.zero_()
        # print(train_each_loss)
    train_all_loss1.append(train_l) # 添加损失值到列表中
    with torch.no_grad():
        test_loss = 0
        for data, labels in traindataloader1:
            pred = model1.forward(data)
            test_each_loss = mse(pred, labels)
            test_loss += test_each_loss.item()
        test_all_loss1.append(test_loss)
    if epoch==0 or (epoch+1) % 4 == 0:
        print('epoch: %d | train loss:%.5f | test loss:%.5f'%(epoch+1,train_all_loss1[-1],test_all_loss1[-1]))

endtime1 = time.time()
print("手动实现前馈网络-回归实验 %d轮 总用时: %.3fs"%(epochs,endtime1-begintime1))

# 二分类任务
data_num2, train_num2, test_num2 =10000, 7000, 3000
# 第一个数据集 符合均值为 0.5 标准差为1 得分布
featuresA = torch.normal(mean=0.5, std=1, size=(data_num2, 200), dtype=torch.float32)
labelsA = torch.ones(data_num2)
# 第二个数据集 符合均值为 -0.5 标准差为1的分布
featuresB = torch.normal(mean=-0.5, std=1, size=(data_num2, 200), dtype=torch.float32)
labelsB = torch.zeros(data_num2)

# 构建训练数据集
train_features2 = torch.cat((featuresA[:train_num2], featuresB[:train_num2]), dim=0) 
train_labels2 = torch.cat((labelsA[:train_num2], labelsB[:train_num2]), dim=-1) 
# 构建测试数据集
test_features2 = torch.cat((featuresA[train_num2:], featuresB[train_num2:]), dim=0)  
test_labels2 = torch.cat((labelsB[train_num2:], labelsB[train_num2:]), dim=-1) 
batch_size = 128
# Build the training and testing dataset
traindataset2 = torch.utils.data.TensorDataset(train_features2, train_labels2)
testdataset2 = torch.utils.data.TensorDataset(test_features2, test_labels2)
traindataloader2 = torch.utils.data.DataLoader(dataset=traindataset2,batch_size=batch_size,shuffle=True)
testdataloader2 = torch.utils.data.DataLoader(dataset=testdataset2,batch_size=batch_size,shuffle=True)

from torch.nn.functional import binary_cross_entropy
from torch.nn import CrossEntropyLoss
class Net2():
    def __init__(self):
        # 设置隐藏层和输出层的节点数
        num_inputs, num_hiddens, num_outputs = 200, 256, 1
        w_1 = torch.tensor(np.random.normal(0, 0.01, (num_hiddens, num_inputs)), dtype=torch.float32,
                           requires_grad=True)
        b_1 = torch.zeros(num_hiddens, dtype=torch.float32, requires_grad=True)
        w_2 = torch.tensor(np.random.normal(0, 0.01, (num_outputs, num_hiddens)), dtype=torch.float32,
                           requires_grad=True)
        b_2 = torch.zeros(num_outputs, dtype=torch.float32, requires_grad=True)
        self.params = [w_1, b_1, w_2, b_2]

        # 定义模型结构
        self.input_layer = lambda x: x.view(x.shape[0], -1)
        self.hidden_layer = lambda x: self.my_relu(torch.matmul(x, w_1.t()) + b_1)
        self.output_layer = lambda x: torch.matmul(x, w_2.t()) + b_2
        self.fn_logistic = self.logistic

    def my_relu(self, x):
        return torch.max(input=x, other=torch.tensor(0.0))

    def logistic(self, x):  # 定义logistic函数
        x = 1.0 / (1.0 + torch.exp(-x))
        return x

    # 定义前向传播
    def forward(self, x):
        x = self.input_layer(x)
        x = self.my_relu(self.hidden_layer(x))
        x = self.fn_logistic(self.output_layer(x))
        return x


# 训练
model2 = Net2()
lr = 0.005  # 学习率
epochs = 40  # 训练轮数
train_all_loss2 = []  # 记录训练集上得loss变化
test_all_loss2 = []  # 记录测试集上的loss变化
train_Acc12, test_Acc12 = [], []
begintime2 = time.time()
for epoch in range(epochs):
    train_l, train_epoch_count = 0, 0
    for data, labels in traindataloader2:
        pred = model2.forward(data)
        train_each_loss = binary_cross_entropy(pred.view(-1), labels.view(-1))  # 计算每次的损失值
        train_l += train_each_loss.item()
        train_each_loss.backward()  # 反向传播
        SGD(model2.params, lr)  # 使用随机梯度下降迭代模型参数
        # 梯度清零
        for param in model2.params:
            param.grad.data.zero_()
        # print(train_each_loss)
        train_epoch_count += (pred.argmax(dim=1) == labels).sum()
    train_Acc12.append((train_epoch_count/len(traindataset2)).item())
    train_all_loss2.append(train_l)  # 添加损失值到列表中
    with torch.no_grad():
        test_l, test_epoch_count = 0, 0
        for data, labels in testdataloader2:
            pred = model2.forward(data)
            test_each_loss = binary_cross_entropy(pred.view(-1), labels.view(-1))
            test_l += test_each_loss.item()
            train_epoch_count += (pred.argmax(dim=1) == labels).sum()
        test_Acc12.append((test_epoch_count/len(testdataset2)).item())
        test_all_loss2.append(test_l)
    if epoch == 0 or (epoch + 1) % 4 == 0:
        print('epoch: %d | train loss:%.5f | test loss:%.5f | train acc:%.5f | test acc:%.5f'  % (epoch + 1, train_all_loss2[-1], test_all_loss2[-1], train_Acc12[-1], test_Acc12[-1]))
endtime2 = time.time()
print("手动实现前馈网络-二分类实验 %d轮 总用时: %.3f" % (epochs, endtime2 - begintime2))

# 多分类任务
mnist_train3 = torchvision.datasets.FashionMNIST(root='./FashionMNIST', train=True, download=True, transform=transforms.ToTensor())
mnist_test3 = torchvision.datasets.FashionMNIST(root='./FashionMNIST', train=False, download=True, transform=transforms.ToTensor())
batch_size = 256
train_iter3 = torch.utils.data.DataLoader(mnist_train3, batch_size=batch_size, shuffle=True, num_workers=0)
test_iter3 = torch.utils.data.DataLoader(mnist_test3, batch_size=batch_size, shuffle=False, num_workers=0)

traindataset3 = torchvision.datasets.FashionMNIST(root='E:\\DataSet\\FashionMNIST\\Train',
                                                  train=True,
                                                  download=True,
                                                  transform=transforms.ToTensor())
testdataset3 = torchvision.datasets.FashionMNIST(root='E:\\DataSet\\FashionMNIST\\Test',
                                                 train=False,
                                                 download=True,
                                                 transform=transforms.ToTensor())
traindataloader3 = torch.utils.data.DataLoader(traindataset3, batch_size=batch_size, shuffle=True)
testdataloader3 = torch.utils.data.DataLoader(testdataset3, batch_size=batch_size, shuffle=False)
# 定义自己的前馈神经网络
class MyNet3():
    def __init__(self):
        # 设置隐藏层和输出层的节点数
        num_inputs, num_hiddens, num_outputs = 28 * 28, 256, 10  # 十分类问题
        w_1 = torch.tensor(np.random.normal(0, 0.01, (num_hiddens, num_inputs)), dtype=torch.float32,
                           requires_grad=True)
        b_1 = torch.zeros(num_hiddens, dtype=torch.float32, requires_grad=True)
        w_2 = torch.tensor(np.random.normal(0, 0.01, (num_outputs, num_hiddens)), dtype=torch.float32,
                           requires_grad=True)
        b_2 = torch.zeros(num_outputs, dtype=torch.float32, requires_grad=True)
        self.params = [w_1, b_1, w_2, b_2]

        # 定义模型结构
        self.input_layer = lambda x: x.view(x.shape[0], -1)
        self.hidden_layer = lambda x: self.my_relu(torch.matmul(x, w_1.t()) + b_1)
        self.output_layer = lambda x: torch.matmul(x, w_2.t()) + b_2

    def my_relu(self, x):
        return torch.max(input=x, other=torch.tensor(0.0))

    # 定义前向传播
    def forward(self, x):
        x = self.input_layer(x)
        x = self.hidden_layer(x)
        x = self.output_layer(x)
        return x


def mySGD(params, lr, batchsize):
    for param in params:
        param.data -= lr * param.grad / batchsize

# 训练
model3 = MyNet3()  # logistics模型
criterion = my_cross_entropy_loss  # 损失函数
lr = 0.15  # 学习率
epochs = 40  # 训练轮数
train_all_loss3 = []  # 记录训练集上得loss变化
test_all_loss3 = []  # 记录测试集上的loss变化
train_ACC13, test_ACC13 = [], [] # 记录正确的个数
begintime3 = time.time()
for epoch in range(epochs):
    train_l,train_acc_num = 0, 0
    for data, labels in traindataloader3:
        pred = model3.forward(data)
        train_each_loss = criterion(pred, labels)  # 计算每次的损失值
        train_l += train_each_loss.item()
        train_each_loss.backward()  # 反向传播
        mySGD(model3.params, lr, 128)  # 使用小批量随机梯度下降迭代模型参数
        # 梯度清零
        train_acc_num += (pred.argmax(dim=1)==labels).sum().item()
        for param in model3.params:
            param.grad.data.zero_()
        # print(train_each_loss)
    train_all_loss3.append(train_l)  # 添加损失值到列表中
    train_ACC13.append(train_acc_num / len(traindataset3)) # 添加准确率到列表中
    with torch.no_grad():
        test_l, test_acc_num = 0, 0
        for data, labels in testdataloader3:
            pred = model3.forward(data)
            test_each_loss = criterion(pred, labels)
            test_l += test_each_loss.item()
            test_acc_num += (pred.argmax(dim=1)==labels).sum().item()
        test_all_loss3.append(test_l)
        test_ACC13.append(test_acc_num / len(testdataset3))   # # 添加准确率到列表中
    if epoch == 0 or (epoch + 1) % 4 == 0:
        print('epoch: %d | train loss:%.5f | test loss:%.5f | train acc: %.2f | test acc: %.2f'
              % (epoch + 1, train_l, test_l, train_ACC13[-1],test_ACC13[-1]))
endtime3 = time.time()
print("手动实现前馈网络-多分类实验 %d轮 总用时: %.3f" % (epochs, endtime3 - begintime3))

# 结果分析
def picture(name, trainl, testl, type='Loss'):
    plt.rcParams["font.sans-serif"]=["SimHei"] #设置字体
    plt.rcParams["axes.unicode_minus"]=False #该语句解决图像中的“-”负号的乱码问题
    plt.title(name) # 命名
    plt.plot(trainl, c='g', label='Train '+ type)
    plt.plot(testl, c='r', label='Test '+type)
    plt.xlabel('Epoch')
    plt.ylabel('Loss')
    plt.legend()
    plt.grid(True)


plt.figure(figsize=(12,3))
plt.title('Loss')
plt.subplot(131)
picture('前馈网络-回归-损失曲线',train_all_loss1,test_all_loss1)
plt.subplot(132)
picture('前馈网络-二分类-损失曲线',train_all_loss2,test_all_loss2)
plt.subplot(133)
picture('前馈网络-多分类-损失曲线',train_all_loss3,test_all_loss3)
plt.show()

# 绘制表格
plt.figure(figsize=(8, 3))
plt.subplot(121)
picture('前馈网络-二分类-准确度',train_Acc12,test_Acc12,type='ACC')
plt.subplot(122)
picture('前馈网络-多分类—准确度', train_ACC13,test_ACC13, type='ACC')
plt.show()

二、利用torch.nn实现前馈神经网络解决回归、二分类、多分类任务

2.1 任务内容

从训练时间、预测精度、Loss变化等角度分析实验结果（最好使用图表展示）

2.2 任务思路及代码

from torch.nn import MSELoss
from torch.optim import SGD
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# 回归任务
class MyNet21(nn.Module):
    def __init__(self):
        super(MyNet21, self).__init__()
        # 设置隐藏层和输出层的节点数
        num_inputs, num_hiddens, num_outputs = 500, 256, 1
        # 定义模型结构
        self.input_layer = nn.Flatten()
        self.hidden_layer = nn.Linear(num_inputs, num_hiddens)
        self.output_layer = nn.Linear(num_hiddens, num_outputs)
        self.relu = nn.ReLU()

    # 定义前向传播
    def forward(self, x):
        x = self.input_layer(x)
        x = self.relu(self.hidden_layer(x))
        x = self.output_layer(x)
        return x

# 训练
model21 = MyNet21()  # logistics模型
model21 = model21.to(device)
print(model21)
criterion = MSELoss()  # 损失函数
criterion = criterion.to(device)
optimizer = SGD(model21.parameters(), lr=0.1)  # 优化函数
epochs = 40  # 训练轮数
train_all_loss21 = []  # 记录训练集上得loss变化
test_all_loss21 = []  # 记录测试集上的loss变化
begintime21 = time.time()
for epoch in range(epochs):
    train_l = 0
    for data, labels in traindataloader1:
        data, labels = data.to(device=device), labels.to(device)
        pred = model21(data)
        train_each_loss = criterion(pred.view(-1, 1), labels.view(-1, 1))  # 计算每次的损失值
        optimizer.zero_grad()  # 梯度清零
        train_each_loss.backward()  # 反向传播
        optimizer.step()  # 梯度更新
        train_l += train_each_loss.item()
    train_all_loss21.append(train_l)  # 添加损失值到列表中
    with torch.no_grad():
        test_loss = 0
        for data, labels in testdataloader1:
            data, labels = data.to(device), labels.to(device)
            pred = model21(data)
            test_each_loss = criterion(pred,labels)
            test_loss += test_each_loss.item()
        test_all_loss21.append(test_loss)
    if epoch == 0 or (epoch + 1) % 10 == 0:
        print('epoch: %d | train loss:%.5f | test loss:%.5f' % (epoch + 1, train_all_loss21[-1], test_all_loss21[-1]))
endtime21 = time.time()
print("torch.nn实现前馈网络-回归实验 %d轮 总用时: %.3fs" % (epochs, endtime21 - begintime21))

class MyNet22(nn.Module):
    def __init__(self):
        super(MyNet22, self).__init__()
        # 设置隐藏层和输出层的节点数
        num_inputs, num_hiddens, num_outputs = 200, 256, 1
        # 定义模型结构
        self.input_layer = nn.Flatten()
        self.hidden_layer = nn.Linear(num_inputs, num_hiddens)
        self.output_layer = nn.Linear(num_hiddens, num_outputs)
        self.relu = nn.ReLU()

    def logistic(self, x):  # 定义logistic函数
        x = 1.0 / (1.0 + torch.exp(-x))
        return x
    # 定义前向传播
    def forward(self, x):
        x = self.input_layer(x)
        x = self.relu(self.hidden_layer(x))
        x = self.logistic(self.output_layer(x))
        return x

# 训练
model22 = MyNet22()  # logistics模型
model22 = model22.to(device)
print(model22)
optimizer = SGD(model22.parameters(), lr=0.001)  # 优化函数
epochs = 40  # 训练轮数
train_all_loss22 = []  # 记录训练集上得loss变化
test_all_loss22 = []  # 记录测试集上的loss变化
train_ACC22, test_ACC22 = [], []
begintime22 = time.time()
for epoch in range(epochs):
    train_l, train_epoch_count, test_epoch_count = 0, 0, 0 # 每一轮的训练损失值 训练集正确个数 测试集正确个数
    for data, labels in traindataloader2:
        data, labels = data.to(device), labels.to(device)
        pred = model22(data)
        train_each_loss = binary_cross_entropy(pred.view(-1), labels.view(-1))  # 计算每次的损失值
        optimizer.zero_grad()  # 梯度清零
        train_each_loss.backward()  # 反向传播
        optimizer.step()  # 梯度更新
        train_l += train_each_loss.item()
        pred = torch.tensor(np.where(pred.cpu()>0.5, 1, 0))  # 大于 0.5时候，预测标签为 1 否则为0
        each_count = (pred.view(-1) == labels.cpu()).sum() # 每一个batchsize的正确个数
        train_epoch_count += each_count # 计算每个epoch上的正确个数
    train_ACC22.append(train_epoch_count / len(traindataset2))
    train_all_loss22.append(train_l)  # 添加损失值到列表中
    with torch.no_grad():
        test_loss, each_count = 0, 0
        for data, labels in testdataloader2:
            data, labels = data.to(device), labels.to(device)
            pred = model22(data)
            test_each_loss = binary_cross_entropy(pred.view(-1),labels)
            test_loss += test_each_loss.item()
            # .cpu 为转换到cpu上计算
            pred = torch.tensor(np.where(pred.cpu() > 0.5, 1, 0))
            each_count = (pred.view(-1)==labels.cpu().view(-1)).sum()
            test_epoch_count += each_count
        test_all_loss22.append(test_loss)
        test_ACC22.append(test_epoch_count / len(testdataset2))
    if epoch == 0 or (epoch + 1) % 4 == 0:
        print('epoch: %d | train loss:%.5f test loss:%.5f | train acc:%.5f | test acc:%.5f' % (epoch + 1, train_all_loss22[-1], 
                                                                                               test_all_loss22[-1], train_ACC22[-1], test_ACC22[-1]))

endtime22 = time.time()
print("torch.nn实现前馈网络-二分类实验 %d轮 总用时: %.3fs" % (epochs, endtime22 - begintime22))

from torch.nn import CrossEntropyLoss
# 定义自己的前馈神经网络
class MyNet23(nn.Module):

    def __init__(self,num_hiddenlayer=1, num_inputs=28*28,num_hiddens=[256],num_outs=10,act='relu'):
        super(MyNet23, self).__init__()
        # 设置隐藏层和输出层的节点数
        self.num_inputs, self.num_hiddens, self.num_outputs = num_inputs,num_hiddens,num_outs # 十分类问题

        # 定义模型结构
        self.input_layer = nn.Flatten()
        # 若只有一层隐藏层
        if num_hiddenlayer ==1:
            self.hidden_layers = nn.Linear(self.num_inputs,self.num_hiddens[-1])
        else: # 若有多个隐藏层
            self.hidden_layers = nn.Sequential()
            self.hidden_layers.add_module("hidden_layer1", nn.Linear(self.num_inputs,self.num_hiddens[0]))
            for i in range(0,num_hiddenlayer-1):
                name = str('hidden_layer'+str(i+2))
                self.hidden_layers.add_module(name, nn.Linear(self.num_hiddens[i],self.num_hiddens[i+1]))
        self.output_layer = nn.Linear(self.num_hiddens[-1], self.num_outputs)
        # 指代需要使用什么样子的激活函数
        if act == 'relu':
            self.act = nn.ReLU()
        elif act == 'sigmoid':
            self.act = nn.Sigmoid()
        elif act == 'tanh':
            self.act = nn.Tanh()
        elif act == 'elu':
            self.act = nn.ELU()
        print(f'本次使用的激活函数为 {act}')
    def logistic(self, x):  # 定义logistic函数
        x = 1.0 / (1.0 + torch.exp(-x))
        return x
    # 定义前向传播
    def forward(self, x):
        x = self.input_layer(x)
        x = self.act(self.hidden_layers(x))
        x = self.output_layer(x)
        return x

# 训练
# 使用默认的参数即： num_inputs=28*28,num_hiddens=256,num_outs=10,act='relu'
model23 = MyNet23()  
model23 = model23.to(device)

# 将训练过程定义为一个函数，方便实验三和实验四调用
def train_and_test(model=model23):
    MyModel = model
    print(MyModel)
    optimizer = SGD(MyModel.parameters(), lr=0.01)  # 优化函数
    epochs = 40  # 训练轮数
    criterion = CrossEntropyLoss() # 损失函数
    train_all_loss23 = []  # 记录训练集上得loss变化
    test_all_loss23 = []  # 记录测试集上的loss变化
    train_ACC23, test_ACC23 = [], []
    begintime23 = time.time()
    for epoch in range(epochs):
        train_l, train_epoch_count, test_epoch_count = 0, 0, 0
        for data, labels in traindataloader3:
            data, labels = data.to(device), labels.to(device)
            pred = MyModel(data)
            train_each_loss = criterion(pred, labels.view(-1))  # 计算每次的损失值
            optimizer.zero_grad()  # 梯度清零
            train_each_loss.backward()  # 反向传播
            optimizer.step()  # 梯度更新
            train_l += train_each_loss.item()
            train_epoch_count += (pred.argmax(dim=1)==labels).sum()
        train_ACC23.append(train_epoch_count.cpu()/len(traindataset3))
        train_all_loss23.append(train_l)  # 添加损失值到列表中
        with torch.no_grad():
            test_loss, test_epoch_count= 0, 0
            for data, labels in testdataloader3:
                data, labels = data.to(device), labels.to(device)
                pred = MyModel(data)
                test_each_loss = criterion(pred,labels)
                test_loss += test_each_loss.item()
                test_epoch_count += (pred.argmax(dim=1)==labels).sum()
            test_all_loss23.append(test_loss)
            test_ACC23.append(test_epoch_count.cpu()/len(testdataset3))
        if epoch == 0 or (epoch + 1) % 4 == 0:
            print('epoch: %d | train loss:%.5f | test loss:%.5f | train acc:%5f test acc:%.5f:' % (epoch + 1, train_all_loss23[-1], test_all_loss23[-1],
                                                                                                                     train_ACC23[-1],test_ACC23[-1]))
    endtime23 = time.time()
    print("torch.nn实现前馈网络-多分类任务 %d轮 总用时: %.3fs" % (epochs, endtime23 - begintime23))
    # 返回训练集和测试集上的 损失值 与 准确率
    return train_all_loss23,test_all_loss23,train_ACC23,test_ACC23
    
train_all_loss23,test_all_loss23,train_ACC23,test_ACC23 = train_and_test(model=model23)

三、在多分类任务中使用至少三种不同的激活函数

3.1 任务内容

使用不同的激活函数，进行对比实验并分析实验结果

3.2 任务思路及代码

# 使用实验二中多分类的模型定义其激活函数为 Tanh
model31 = MyNet23(1,28*28,[256],10,act='tanh') 
model31 = model31.to(device)

train_all_loss31,test_all_loss31,train_ACC31,test_ACC31 = train_and_test(model=model31)

# 使用实验二中多分类的模型定义其激活函数为 Sigmoid
model32 = MyNet23(1,28*28,[256],10,act='sigmoid') 
model32 = model32.to(device)

train_all_loss32,test_all_loss32,train_ACC32,test_ACC32 = train_and_test(model=model32)

# 使用实验二中多分类的模型定义其激活函数为 ELU
model33 = MyNet23(1,28*28,[256],10,act='elu') 
model33 = model33.to(device) 

train_all_loss33,test_all_loss33,train_ACC33,test_ACC33 = train_and_test(model=model33)

def Plot3(datalist,title='1',ylabel='Loss',flag='act'):
    plt.rcParams["font.sans-serif"]=["SimHei"] #设置字体
    plt.rcParams["axes.unicode_minus"]=False #该语句解决图像中的“-”负号的乱码问题
    plt.title(title)
    plt.xlabel('Epoch')
    plt.ylabel(ylabel)
    plt.plot(datalist[0],label='Tanh' if flag=='act' else '[128]')
    plt.plot(datalist[1],label='Sigmoid' if flag=='act' else '[512 256]')
    plt.plot(datalist[2],label='ELu' if flag=='act' else '[512 256 128 64]')
    plt.plot(datalist[3],label='Relu' if flag=='act' else '[256]')
    plt.legend()
    plt.grid(True)

plt.figure(figsize=(16,3))
plt.subplot(141)
Plot3([train_all_loss31,train_all_loss32,train_all_loss33,train_all_loss23],title='Train_Loss')
plt.subplot(142)
Plot3([test_all_loss31,test_all_loss32,test_all_loss33,test_all_loss23],title='Test_Loss')
plt.subplot(143)
Plot3([train_ACC31,train_ACC32,train_ACC33,train_ACC23],title='Train_ACC')
plt.subplot(144)
Plot3([test_ACC31,test_ACC32,test_ACC33,test_ACC23],title='Test_ACC')
plt.show()

四、在多分类任务中使用至少三种不同的激活函数

4.1 任务内容

使用不同的隐藏层层数和隐藏单元个数，进行对比实验并分析实验结果

# 使用实验二中多分类的模型  一个隐藏层，神经元个数为[128]
model41 = MyNet23(1,28*28,[128],10,act='relu') 
model41 = model41.to(device) 

train_all_loss41,test_all_loss41,train_ACC41,test_ACC41 = train_and_test(model=model41)

# 使用实验二中多分类的模型 两个隐藏层，神经元个数为[512，256]
model42 = MyNet23(2,28*28,[512,256],10,act='relu') 
model42 = model42.to(device)

train_all_loss42,test_all_loss42,train_ACC42,test_ACC42 = train_and_test(model=model42)

# 使用实验二中多分类的模型  四个隐藏层，神经元个数为[512，256，128，64]
model43 = MyNet23(3,28*28,[512,256,128],10,act='relu') 
model43 = model43.to(device) 

train_all_loss43,test_all_loss43,train_ACC43,test_ACC43 = train_and_test(model=model43)

plt.figure(figsize=(16,3))
plt.subplot(141)
Plot3([train_all_loss41,train_all_loss42,train_all_loss43,train_all_loss23],title='Train_Loss',flag='hidden')
plt.subplot(142)
Plot3([test_all_loss41,test_all_loss42,test_all_loss43,test_all_loss23],title='Test_Loss',flag='hidden')
plt.subplot(143)
Plot3([train_ACC41,train_ACC42,train_ACC43,train_ACC23],title='Train_ACC',flag='hidden')
plt.subplot(144)
Plot3([test_ACC41,test_ACC42,test_ACC43,test_ACC23],title='Test_ACC', flag='hidden')
plt.show()

实验结果分析

1. 从训练时间大致可以看出：隐藏层数越多，隐藏神经元个数越多，训练成本越高，所需要的时间越久。
2. 从准确率来看，准确率越高，可能会有相反的效果，并不是隐藏层数越多，隐藏神经元个数越多。更多的隐藏层和隐藏神经元个数，可能会导致模型的过拟合现象，导致在训练集上准确率很高，但在测试集上准确率很低。