第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

1.数据预处理

1.1 设置GPU

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

device(type=‘cpu’)

1.2 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
plt.rcParams['savefig.dpi'] = 500 #图片像素
plt.rcParams['figure.dpi']  = 500 #分辨率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('dia.xls')
DataFrame.head()

	卡号	性别	年龄	高密度脂蛋白胆固醇	低密度脂蛋白胆固醇	极低密度脂蛋白胆固醇	甘油三酯	总胆固醇	脉搏	舒张压	高血压史	尿素氮	尿酸	肌酐	体重检查结果	是否糖尿病
0	18054421	0	38	1.25	2.99	1.07	0.64	5.31	83	83	0	4.99	243.3	50	1	0
1	18054422	0	31	1.15	1.99	0.84	0.50	3.98	85	63	0	4.72	391.0	47	1	0
2	18054423	0	27	1.29	2.21	0.69	0.60	4.19	73	61	0	5.87	325.7	51	1	0
3	18054424	0	33	0.93	2.01	0.66	0.84	3.60	83	60	0	2.40	203.2	40	2	0
4	18054425	0	36	1.17	2.83	0.83	0.73	4.83	85	67	0	4.09	236.8	43	0	0

DataFrame.shape

(1006, 16)

1.3 数据检查

# 查看数据是否有缺失值
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('数据集的重复值为:'f'{DataFrame.duplicated().sum()}')

数据重复值---------------------------------
数据集的重复值为:0

2. 数据分析

2.1 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()

2.2 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)

3. LSTM模型

3.1 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，故而在 X 中去掉该字段
X = DataFrame.drop(['是否糖尿病','高密度脂蛋白胆固醇'],axis=1)
y = DataFrame['是否糖尿病']# 数据集标准化处理
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)
# 维度扩增使其符合LSTM模型可接受shape
train_X = train_X.unsqueeze(1)
test_X  = test_X.unsqueeze(1)
train_X.shape, train_y.shape

(torch.Size([804, 1, 13]), torch.Size([804]))

3.2 数据集构建

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)

3.3 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)def forward(self, x):out, hidden1 = self.lstm0(x) out, _ = self.lstm1(out, hidden1) out    = out[:, -1, :]  # 只取最后一个时间步的输出out    = self.fc0(out) return out   model = model_lstm().to(device)
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)
)

4. 训练模型

4.1 定义训练函数

# 训练循环
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, (size/batch_size，向上取整)train_loss, train_acc = 0, 0  # 初始化训练损失和正确率for X, y in dataloader:  # 获取图片及其标签X, y = X.to(device), y.to(device)# 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，targets为真实值，计算二者差值即为损失# 反向传播optimizer.zero_grad()  # grad属性归零loss.backward()        # 反向传播optimizer.step()       # 每一步自动更新# 记录acc与losstrain_acc  += (pred.argmax(1) == y).type(torch.float).sum().item()train_loss += loss.item()train_acc  /= sizetrain_loss /= num_batchesreturn train_acc, train_loss

4.2 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，向上取整)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

4.3 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     = 30train_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:43.8%, Train_loss:0.693, Test_acc:48.0%, Test_loss:0.682, Lr:1.00E-04
Epoch: 2, Train_acc:52.6%, Train_loss:0.684, Test_acc:62.4%, Test_loss:0.676, Lr:1.00E-04
Epoch: 3, Train_acc:68.4%, Train_loss:0.674, Test_acc:69.8%, Test_loss:0.669, Lr:1.00E-04
Epoch: 4, Train_acc:72.9%, Train_loss:0.662, Test_acc:73.8%, Test_loss:0.661, Lr:1.00E-04
Epoch: 5, Train_acc:76.1%, Train_loss:0.648, Test_acc:74.3%, Test_loss:0.651, Lr:1.00E-04
Epoch: 6, Train_acc:76.4%, Train_loss:0.631, Test_acc:73.8%, Test_loss:0.639, Lr:1.00E-04
Epoch: 7, Train_acc:76.1%, Train_loss:0.611, Test_acc:74.3%, Test_loss:0.625, Lr:1.00E-04
Epoch: 8, Train_acc:76.0%, Train_loss:0.588, Test_acc:75.2%, Test_loss:0.610, Lr:1.00E-04
Epoch: 9, Train_acc:75.0%, Train_loss:0.564, Test_acc:75.2%, Test_loss:0.595, Lr:1.00E-04
Epoch:10, Train_acc:75.0%, Train_loss:0.541, Test_acc:75.2%, Test_loss:0.581, Lr:1.00E-04
Epoch:11, Train_acc:75.2%, Train_loss:0.521, Test_acc:75.2%, Test_loss:0.569, Lr:1.00E-04
Epoch:12, Train_acc:75.7%, Train_loss:0.504, Test_acc:75.7%, Test_loss:0.559, Lr:1.00E-04
Epoch:13, Train_acc:75.7%, Train_loss:0.491, Test_acc:75.7%, Test_loss:0.550, Lr:1.00E-04
Epoch:14, Train_acc:75.6%, Train_loss:0.480, Test_acc:76.7%, Test_loss:0.543, Lr:1.00E-04
Epoch:15, Train_acc:75.7%, Train_loss:0.472, Test_acc:76.2%, Test_loss:0.535, Lr:1.00E-04
Epoch:16, Train_acc:76.7%, Train_loss:0.465, Test_acc:76.2%, Test_loss:0.529, Lr:1.00E-04
Epoch:17, Train_acc:77.4%, Train_loss:0.459, Test_acc:76.7%, Test_loss:0.522, Lr:1.00E-04
Epoch:18, Train_acc:77.9%, Train_loss:0.454, Test_acc:77.2%, Test_loss:0.516, Lr:1.00E-04
Epoch:19, Train_acc:78.4%, Train_loss:0.450, Test_acc:77.7%, Test_loss:0.511, Lr:1.00E-04
Epoch:20, Train_acc:78.2%, Train_loss:0.446, Test_acc:77.2%, Test_loss:0.506, Lr:1.00E-04
Epoch:21, Train_acc:78.2%, Train_loss:0.442, Test_acc:77.2%, Test_loss:0.501, Lr:1.00E-04
Epoch:22, Train_acc:78.6%, Train_loss:0.439, Test_acc:77.2%, Test_loss:0.496, Lr:1.00E-04
Epoch:23, Train_acc:78.9%, Train_loss:0.436, Test_acc:77.2%, Test_loss:0.492, Lr:1.00E-04
Epoch:24, Train_acc:78.9%, Train_loss:0.433, Test_acc:77.7%, Test_loss:0.488, Lr:1.00E-04
Epoch:25, Train_acc:79.2%, Train_loss:0.430, Test_acc:77.7%, Test_loss:0.484, Lr:1.00E-04
Epoch:26, Train_acc:79.2%, Train_loss:0.427, Test_acc:78.2%, Test_loss:0.481, Lr:1.00E-04
Epoch:27, Train_acc:79.4%, Train_loss:0.425, Test_acc:79.2%, Test_loss:0.477, Lr:1.00E-04
Epoch:28, Train_acc:79.4%, Train_loss:0.423, Test_acc:79.2%, Test_loss:0.474, Lr:1.00E-04
Epoch:29, Train_acc:79.5%, Train_loss:0.421, Test_acc:79.2%, Test_loss:0.471, Lr:1.00E-04
Epoch:30, Train_acc:79.6%, Train_loss:0.418, Test_acc:79.7%, Test_loss:0.467, Lr:1.00E-04
==================== Done ====================

5. 模型评估

5.1 Loss与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() # 获取当前时间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) # 打卡请带上时间戳，否则代码截图无效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()

6. 总结

本周主要实现了实现了对于上一次糖尿病预测模型的优化。通过实践，更加深入地了解了LSTM模型的相关优化。