导入所需库

import torch import torch.nn as nn import pandas as pd from sklearn import preprocessing import numpy as np

读取Excel表格

data = pd.read_excel('C:\Users\lenovo\Desktop\HIV\DNN神经网络测试\data1.xlsx') x_data = data.iloc[:, 1:].values.astype(float) # 去掉第一列和第一行 y_data = data.iloc[:, 0].values.astype(float)

数据标准化

scaler = preprocessing.StandardScaler().fit(x_data) x_data = scaler.transform(x_data)

定义第一个模型

class Model1(nn.Module): def init(self, input_size, hidden_size, num_classes): super(Model1, self).init() self.fc1 = nn.Linear(input_size, hidden_size) self.relu = nn.ReLU() self.dropout = nn.Dropout(p=0.5) self.fc2 = nn.Linear(hidden_size, num_classes)

def forward(self, x):
    out = self.fc1(x)
    out = self.relu(out)
    out = self.dropout(out)
    out = self.fc2(out)
    return out

定义第二个模型

class Model2(nn.Module): def init(self, input_size, hidden_size, num_classes): super(Model2, self).init() self.fc1 = nn.Linear(input_size, hidden_size) self.relu = nn.ReLU() self.dropout = nn.Dropout(p=0.5) self.fc2 = nn.Linear(hidden_size, num_classes) self.sigmoid = nn.Sigmoid()

def forward(self, x):
    out = self.fc1(x)
    out = self.relu(out)
    out = self.dropout(out)
    out = self.fc2(out)
    out = self.sigmoid(out)
    return out

定义模型参数

input_size = x_data.shape[1] hidden_size = 256 num_classes1 = 4 num_classes2 = 1 learning_rate = 0.001 num_epochs = 1000

初始化模型

model1 = Model1(input_size, hidden_size, num_classes1) model2 = Model2(num_classes1, hidden_size, num_classes2)

定义损失函数和优化器

criterion1 = nn.CrossEntropyLoss() criterion2 = nn.BCELoss() optimizer1 = torch.optim.Adam(model1.parameters(), lr=learning_rate) optimizer2 = torch.optim.Adam(model2.parameters(), lr=learning_rate)

将数据转换成Tensor类型

x_data = torch.Tensor(x_data) y_data = torch.LongTensor(y_data)

训练第一个模型

for epoch in range(num_epochs): # 向前传播 outputs = model1(x_data) loss = criterion1(outputs, y_data)

# 反向传播和优化
optimizer1.zero_grad()
loss.backward()
optimizer1.step()

# 计算准确率
_, predicted = torch.max(outputs.data, 1)
total = predicted.size(0)
correct = predicted.eq(y_data.data).cpu().sum()
accuracy = 100.0 * correct / total

# 输出结果
if (epoch + 1) % 100 == 0:
    print('Epoch [%d/%d], Loss: %.4f, Accuracy: %.2f %%' % (epoch + 1, num_epochs, loss.item(), accuracy))

将第一个模型的输出作为第二个模型的输入

x_data2 = model1(x_data) y_data2 = y_data.view(-1, 1).float()

训练第二个模型

for epoch in range(num_epochs): # 向前传播 outputs = model2(x_data2) loss = criterion2(outputs, y_data2)

# 反向传播和优化
optimizer2.zero_grad()
loss.backward()
optimizer2.step()

# 计算准确率
predicted = outputs.round()
total = predicted.size(0)
correct = predicted.eq(y_data2).cpu().sum()
accuracy = 100.0 * correct / total

# 输出结果
if (epoch + 1) % 100 == 0:
    print('Epoch [%d/%d], Loss: %.4f, Accuracy: %.2f %%' % (epoch + 1, num_epochs, loss.item(), accuracy))