基于深度神经网络的HIV预测模型

本代码使用深度神经网络 (DNN) 构建了一个 HIV 预测模型。

1. 导入必要的库

import torch
import torch.nn as nn
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.metrics import roc_curve, auc

2. 读取数据

data = pd.read_excel('C:\Users\lenovo\Desktop\HIV\DNN神经网络测试\output_data.xlsx')
x = data.iloc[:, 1:].values
y = data.iloc[:, 0].values

3. 数据归一化

x = (x - x.mean()) / x.std()

4. 将numpy数组转换为张量

x = torch.Tensor(x)
y = torch.Tensor(y)

5. 定义模型

class DNN(nn.Module):
    def __init__(self, input_size, hidden_size, num_classes):
        super(DNN, self).__init__()
        self.fc1 = nn.Linear(input_size, hidden_size)
        self.fc2 = nn.Linear(hidden_size, hidden_size)
        self.fc3 = nn.Linear(hidden_size, hidden_size)
        self.out = nn.Linear(hidden_size, num_classes)
        self.dropout = nn.Dropout(p=0.5)
        self.attention = nn.Sequential(
            nn.Linear(hidden_size, 1),
            nn.Tanh(),
            nn.Softmax(dim=1)
        )

    def forward(self, x):
        out = torch.relu(self.fc1(x))
        out = self.dropout(out)
        out = torch.relu(self.fc2(out))
        out = self.dropout(out)
        out = torch.relu(self.fc3(out))
        out = self.dropout(out)
        attention_weights = self.attention(out)
        out = attention_weights * out
        out = self.out(out)
        return out

6. 定义超参数

input_size = 16
hidden_size = 128
num_classes = 2
learning_rate = 0.001
num_epochs = 100

7. 初始化模型

model = DNN(input_size, hidden_size, num_classes)

8. 定义损失函数和优化器

criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

9. 训练模型

train_loss = []
train_accuracy = []
for epoch in range(num_epochs):
    # 前向传播和反向传播
    outputs = model(x)
    loss = criterion(outputs, y.long())
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

    # 计算准确率
    _, predicted = torch.max(outputs.data, 1)
    total = y.size(0)
    correct = (predicted == y.long()).sum().item()
    accuracy = correct / total
    train_accuracy.append(accuracy)
    train_loss.append(loss.item())

    # 输出训练信息
    print('Epoch [{}/{}], Loss: {:.4f}, Accuracy: {:.2f}%'
          .format(epoch + 1, num_epochs, loss.item(), accuracy * 100))

# 输出每个样本的概率
prob = torch.softmax(outputs, dim=1)[:, 1]
print('每个样本的概率:', prob.detach().numpy().reshape(-1, 1))

# 绘制准确率变化的图
plt.plot(train_accuracy)
plt.title('Accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.show()

# 绘制损失变化的图
plt.plot(train_loss)
plt.title('Loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.show()

# 绘制ROC图
fpr, tpr, threshold = roc_curve(y, prob.detach().numpy())
roc_auc = auc(fpr, tpr)
plt.title('ROC Curve')
plt.plot(fpr, tpr, 'b', label='AUC = %0.2f' % roc_auc)
plt.legend(loc='lower right')
plt.plot([0, 1], [0, 1], 'r--')
plt.xlim([0, 1])
plt.ylim([0, 1])
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show()

10. 读取验证集数据

val_data = pd.read_excel('C:\Users\lenovo\Desktop\HIV\DNN神经网络测试\HIV数据.xlsx')
val_x = val_data.iloc[:, 1:].values
val_y = val_data.iloc[:, 0].values

11. 数据归一化

val_x = torch.Tensor(val_x)
val_y = torch.Tensor(val_y)
val_x = (val_x - x_mean) / x_std

12. 验证模型

model.eval()
val_outputs = model(val_x)
val_loss = criterion(val_outputs, val_y.long())

# 计算准确率
_, val_predicted = torch.max(val_outputs.data, 1)
val_total = val_y.size(0)
val_correct = (val_predicted == val_y.long()).sum().item()
val_accuracy = val_correct / val_total
print('Validation Loss: {:.4f}, Accuracy: {:.2f}%'.format(val_loss.item(), val_accuracy * 100))

# 输出验证集每个样本的概率
val_prob = torch.softmax(val_outputs, dim=1)[:, 1]
print('Validation每个样本的概率:', val_prob.detach().numpy().reshape(-1, 1))

# 绘制验证集ROC图
val_fpr, val_tpr, val_threshold = roc_curve(val_y, val_prob.detach().numpy())
val_roc_auc = auc(val_fpr, val_tpr)
plt.title('Validation ROC Curve')
plt.plot(val_fpr, val_tpr, 'b', label='AUC = %0.2f' % val_roc_auc)
plt.legend(loc='lower right')
plt.plot([0, 1], [0, 1], 'r--')
plt.xlim([0, 1])
plt.ylim([0, 1])
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show()

13. 解决错误

在验证阶段，当进行归一化时，代码出现了错误：TypeError: unsupported operand type(s) for -: 'numpy.ndarray' and 'Tensor'

出现该错误是因为 val_x 是 numpy 数组，而 x_mean 和 x_std 是 PyTorch 张量。为了解决这个问题，需要将 val_x 转换为 PyTorch 张量。

修改后的代码：

# 读取验证集数据
val_data = pd.read_excel('C:\Users\lenovo\Desktop\HIV\DNN神经网络测试\HIV数据.xlsx')
val_x = val_data.iloc[:, 1:].values
val_y = val_data.iloc[:, 0].values

# 将numpy数组转换为张量
val_x = torch.Tensor(val_x)
val_y = torch.Tensor(val_y)

# 数据归一化
val_x = (val_x - x_mean) / x_std

通过将 val_x 转换为 PyTorch 张量，可以解决该错误。

总结

本代码展示了如何使用深度神经网络构建一个 HIV 预测模型，并使用注意力机制来提升模型性能。通过训练和验证，模型取得了不错的效果。