R7周：糖尿病预测模型优化探索

•      🍨 本文为🔗365天深度学习训练营中的学习记录博客
•      🍖 原作者：K同学啊

一、数据预处理

1.设置GPU

import torch.nn.functional as F
import torch.nn as nn
import torch, torchvisiondevice = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device

device(type='cuda')

2.数据导入

import numpy as np
import pandas as pd
import seaborn as sns
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
# 支持中文
plt.rcParams['savefig.dpi'] = 500
plt.rcParams['figure.dpi'] = 500
plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签import warnings
warnings.filterwarnings("ignore")DataFrame = pd.read_excel('F:/jupyter lab/DL-100-days/datasets/diabetes_pre/dia.xls')
DataFrame.head()

​

​

DataFrame.shape

(1006, 16)

 3.数据检查

# 查看是否有缺失值
print("数据缺失值------------------")
print(DataFrame.isnull().sum())

数据缺失值------------------
卡号            0
性别            0
年龄            0
高密度脂蛋白胆固醇     0
低密度脂蛋白胆固醇     0
极低密度脂蛋白胆固醇    0
甘油三酯          0
总胆固醇          0
脉搏            0
舒张压           0
高血压史          0
尿素氮           0
尿酸            0
肌酐            0
体重检查结果        0
是否糖尿病         0
dtype: int64

# 查看数据是否有重复值
print("数据重复值------------------")
print('数据的重复值为：'f'{DataFrame.duplicated().sum()}')

数据重复值------------------
数据的重复值为：0

二、数据分析

1.数据分布分析 

feature_map = {'年龄': '年龄','高密度脂蛋白胆固醇': '高密度脂蛋白胆固醇','低密度脂蛋白胆固醇': '低密度脂蛋白胆固醇','极低密度脂蛋白胆固醇': '极低密度脂蛋白胆固醇','甘油三酯': '甘油三酯','总胆固醇': '总胆固醇','脉搏': '脉搏','舒张压': '舒张压','高血压史': '高血压史','尿素氮': '尿素氮','尿酸': '尿酸','肌酐': '肌酐','体重检查结果': '体重检查结果'
}plt.figure(figsize=(15, 10))
for i, (col, col_name) in enumerate(feature_map.items(), 1):plt.subplot(3, 5, i)sns.boxplot(x=DataFrame['是否糖尿病'], y=DataFrame[col])plt.title(f'{col_name}的箱线图', fontsize=14)plt.ylabel('数值', fontsize=12)plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

​

2. 相关性分析

import plotly
import plotly.express as px#删除列'卡号'
DataFrame.drop(columns=['卡号'], inplace=True)
# 计算各列之间的相关系数
df_corr = DataFrame.corr()#相关矩阵生成函数
def corr_generate(df):fig = px.imshow(df,text_auto=True,aspect="auto",color_continuous_scale='RdBu_r')fig.show()#生成相关矩阵
corr_generate(df_corr)

 ​

三、LSTM模型

1.划分数据集

from sklearn.preprocessing import StandardScaler# '高密度脂蛋白胆固醇'字段与糖尿病负相关，故在X 中去掉该字段
X = DataFrame.drop(['卡号', '是否糖尿病', '高密度脂蛋白胆固醇'], axis=1)
y = DataFrame['是否糖尿病']sc_X = StandardScaler
X = sc_X.fit_transform(X)X = torch.tensor(np.array(X), dtype=torch.float32)
y = torch.tensor(np.array(y), dtype=torch.int64)train_X, test_X, train_y, test_y = train_test_split(X, y, test_size=0.2, random_state=1)train_X.shape, train_y.shape

(torch.Size([804, 13]), torch.Size([804]))

from torch.utils.data import TensorDataset, DataLoadertrain_dl = DataLoader(TensorDataset(train_X, train_y), batch_size=64, shuffle=False)
test_dl = DataLoader(TensorDataset(test_X, test_y), batch_size=64, shuffle=False)

2.定义模型

class model_lstm(nn.Module):def __init__(self):super(model_lstm, self).__init__()self.lstm0 = nn.LSTM(input_size=13, hidden_size=200,num_layers=1, batch_first=True)self.lstm1 = nn.LSTM(input_size=200, hidden_size=200,num_layers=1, batch_first=True)self.fc0 = nn.Linear(200, 2)  # 输出 2 类def forward(self, x):# 如果 x 是 2D 的，转换为 3D 张量，假设 seq_len=1if x.dim() == 2:x = x.unsqueeze(1)  # [batch_size, 1, input_size]# LSTM 处理数据out, (h_n, c_n) = self.lstm0(x)  # 第一层 LSTM# 使用第二个 LSTM，并传递隐藏状态out, (h_n, c_n) = self.lstm1(out, (h_n, c_n))  # 第二层 LSTM# 获取最后一个时间步的输出out = out[:, -1, :]  # 选择序列的最后一个时间步的输出out = self.fc0(out)  # [batch_size, 2]return outmodel = model_lstm().to(device)
print(model)

model_lstm((lstm0): LSTM(13, 200, batch_first=True)(lstm1): LSTM(200, 200, batch_first=True)(fc0): Linear(in_features=200, out_features=2, bias=True)
)

三、训练模型

1.定义训练函数

def train(dataloader, model, loss_fn, optimizer):size = len(dataloader.dataset)  # 训练集的大小num_batches = len(dataloader)  # 批次数目train_loss, train_acc = 0, 0  # 初始化训练损失和正确率model.train()  # 设置模型为训练模式for X, y in dataloader:  # 获取数据和标签# 如果 X 是 2D 的，调整为 3Dif X.dim() == 2:X = X.unsqueeze(1)  # [batch_size, 1, input_size]，即假设 seq_len=1X, y = X.to(device), y.to(device)  # 将数据移动到设备# 计算预测误差pred = model(X)  # 网络输出loss = loss_fn(pred, y)  # 计算网络输出和真实值之间的差距# 反向传播optimizer.zero_grad()  # 清除上一步的梯度loss.backward()  # 反向传播optimizer.step()  # 更新权重# 记录acc与losstrain_acc += (pred.argmax(1) == y).type(torch.float).sum().item()train_loss += loss.item()train_acc /= size  # 平均准确率train_loss /= num_batches  # 平均损失return train_acc, train_loss

2.定义测试函数

def test(dataloader, model, loss_fn):size = len(dataloader.dataset)  # 测试集的大小num_batches = len(dataloader)  # 批次数目, (size/batch_size，向上取test_loss, test_acc = 0, 0# 当不进行训练时，停止梯度更新，节省计算内存消耗with torch.no_grad():for imgs, target in dataloader:imgs, target = imgs.to(device), target.to(device)# 计算losstarget_pred = model(imgs)loss = loss_fn(target_pred, target)test_loss += loss.item()test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()test_acc /= sizetest_loss /= num_batchesreturn test_acc, test_loss

3.训练模型

loss_fn = nn.CrossEntropyLoss()  # 创建损失函数
learn_rate = 1e-4  # 学习率
opt = torch.optim.Adam(model.parameters(), lr=learn_rate)
epochs = 30
train_loss = []
train_acc = []
test_loss = []
test_acc = []
for epoch in range(epochs):model.train()epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, opt)model.eval()epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)train_acc.append(epoch_train_acc)train_loss.append(epoch_train_loss)test_acc.append(epoch_test_acc)test_loss.append(epoch_test_loss)# 获取当前的学习率lr = opt.state_dict()['param_groups'][0]['lr']template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f},Lr:{:.2E}')print(template.format(epoch + 1, epoch_train_acc * 100, epoch_train_loss, epoch_test_acc * 100, epoch_test_loss, lr))print("=" * 20, 'Done', "=" * 20)

Epoch: 1, Train_acc:56.5%, Train_loss:0.688, Test_acc:53.0%, Test_loss:0.704,Lr:1.00E-04
Epoch: 2, Train_acc:56.3%, Train_loss:0.681, Test_acc:53.0%, Test_loss:0.704,Lr:1.00E-04
Epoch: 3, Train_acc:56.3%, Train_loss:0.676, Test_acc:53.0%, Test_loss:0.697,Lr:1.00E-04
Epoch: 4, Train_acc:56.3%, Train_loss:0.670, Test_acc:53.0%, Test_loss:0.690,Lr:1.00E-04
Epoch: 5, Train_acc:56.2%, Train_loss:0.663, Test_acc:54.5%, Test_loss:0.684,Lr:1.00E-04
..........

Epoch:26, Train_acc:76.6%, Train_loss:0.481, Test_acc:71.3%, Test_loss:0.546,Lr:1.00E-04
Epoch:27, Train_acc:76.9%, Train_loss:0.475, Test_acc:71.8%, Test_loss:0.541,Lr:1.00E-04
Epoch:28, Train_acc:77.5%, Train_loss:0.470, Test_acc:71.3%, Test_loss:0.537,Lr:1.00E-04
Epoch:29, Train_acc:77.2%, Train_loss:0.465, Test_acc:71.8%, Test_loss:0.533,Lr:1.00E-04
Epoch:30, Train_acc:77.4%, Train_loss:0.460, Test_acc:70.8%, Test_loss:0.529,Lr:1.00E-04
==================== Done ====================

五、模型评估

1.Loss和Accuracy图

import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore")               #忽略警告信息
plt.rcParams['font.sans-serif']    = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False      # 用来正常显示负号
plt.rcParams['figure.dpi']         = 100        #分辨率from datetime import datetime
current_time = datetime.now()epochs_range = range(epochs)plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.xlabel(current_time)plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

​

六、学习心得

1.本周延续上周的工作，开展了糖尿病预测模型优化探索。加入了相关性分析这个新模块，更加直观地实现了各种因素之间的相关性。

2.从训练结果中可以发现，test_acc有所增长。

3.相较于R6而言，主要修改的地方在于数据集那部分，取消注释了sc_X= StandardScaler()和X= sc_X.fit_transform(X)两行代码。