- 🍨 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、进阶说明
通过矩阵图发现一些无关的特征对预测结果捎有影响，所以去掉一些无用的特征以提升精度
二、代码实现
1.导入库函数
import torch.nn as nn
import torch.nn.functional as F
import torchvision,torch
import plotly
import plotly.express as px
from sklearn.preprocessing import StandardScaler
# 设置硬件设备，如果有GPU则使用，没有则使用cpu
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
import matplotlib.pyplot as plt
#隐藏警告
import warnings
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
2.导入数据
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
warnings.filterwarnings("ignore")
DataFrame=pd.read_excel(r"E:\Code\pytorch_gpu\data\dia.xls")
print(DataFrame.head())
# 查看数据是否有缺失值
print('数据缺失值---------------------------------')
print(DataFrame.isnull().sum())
# 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()}')
卡号 性别 年龄 高密度脂蛋白胆固醇 低密度脂蛋白胆固醇 ... 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 ... 4.99 243.3 50 1 0
1 18054422 0 31 1.15 1.99 ... 4.72 391.0 47 1 0
2 18054423 0 27 1.29 2.21 ... 5.87 325.7 51 1 0
3 18054424 0 33 0.93 2.01 ... 2.40 203.2 40 2 0
4 18054425 0 36 1.17 2.83 ... 4.09 236.8 43 0 0
[5 rows x 16 columns]
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
数据重复值---------------------------------
数据集的重复值为:0
3.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()
4.数据集构建
#删除卡号列
DataFrame.drop(columns=['卡号'], inplace=True)
#计算割裂之间的相关系数
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)
# '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']
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 = train_X.unsqueeze(1)
test_X = test_X.unsqueeze(1)
print(train_X.shape, train_y.shape)#
from torch.utils.data import TensorDataset, DataLoader
train_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)
5.模型构建
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 = self.fc0(out)
return out
model = model_lstm().to(device)
model lstm(
(lstm0):LSTM(13，200，batch first=True)
(1stm1):LSTM(200，200，batch first=True)
(fc0): Linear(in features=200,out features=2,bias=True)
6.构建测试训练
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loss
train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss += loss.item()
train_acc /= size
train_loss /= num_batches
return train_acc, train_loss
7. 构建训练函数
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loss
target_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 /= size
test_loss /= num_batches
return test_acc, test_loss
8.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:56.2%, Train_loss:0.685, Test_acc:52.5%, Test_loss:0.690, Lr:1.00E-04
==================== Done ====================
Epoch: 2, Train_acc:58.5%, Train_loss:0.676, Test_acc:57.4%, Test_loss:0.683, Lr:1.00E-04
==================== Done ====================
Epoch: 3, Train_acc:64.6%, Train_loss:0.665, Test_acc:60.4%, Test_loss:0.675, Lr:1.00E-04
==================== Done ====================
Epoch: 4, Train_acc:66.5%, Train_loss:0.654, Test_acc:62.9%, Test_loss:0.667, Lr:1.00E-04
==================== Done ====================
Epoch: 5, Train_acc:69.5%, Train_loss:0.640, Test_acc:64.9%, Test_loss:0.656, Lr:1.00E-04
==================== Done ====================
Epoch: 6, Train_acc:71.6%, Train_loss:0.624, Test_acc:66.8%, Test_loss:0.644, Lr:1.00E-04
==================== Done ====================
Epoch: 7, Train_acc:73.4%, Train_loss:0.606, Test_acc:70.3%, Test_loss:0.631, Lr:1.00E-04
==================== Done ====================
Epoch: 8, Train_acc:74.5%, Train_loss:0.586, Test_acc:72.8%, Test_loss:0.616, Lr:1.00E-04
==================== Done ====================
Epoch: 9, Train_acc:74.6%, Train_loss:0.564, Test_acc:73.3%, Test_loss:0.601, Lr:1.00E-04
==================== Done ====================
Epoch:10, Train_acc:75.1%, Train_loss:0.543, Test_acc:73.8%, Test_loss:0.587, Lr:1.00E-04
==================== Done ====================
Epoch:11, Train_acc:75.4%, Train_loss:0.523, Test_acc:74.8%, Test_loss:0.574, Lr:1.00E-04
==================== Done ====================
Epoch:12, Train_acc:75.6%, Train_loss:0.506, Test_acc:74.8%, Test_loss:0.562, Lr:1.00E-04
==================== Done ====================
Epoch:13, Train_acc:75.6%, Train_loss:0.492, Test_acc:74.8%, Test_loss:0.552, Lr:1.00E-04
==================== Done ====================
Epoch:14, Train_acc:75.7%, Train_loss:0.481, Test_acc:75.7%, Test_loss:0.543, Lr:1.00E-04
==================== Done ====================
Epoch:15, Train_acc:76.9%, Train_loss:0.472, Test_acc:75.2%, Test_loss:0.535, Lr:1.00E-04
==================== Done ====================
Epoch:16, Train_acc:77.0%, Train_loss:0.464, Test_acc:75.2%, Test_loss:0.528, Lr:1.00E-04
==================== Done ====================
Epoch:17, Train_acc:77.7%, Train_loss:0.458, Test_acc:75.7%, Test_loss:0.522, Lr:1.00E-04
==================== Done ====================
Epoch:18, Train_acc:78.1%, Train_loss:0.452, Test_acc:76.2%, Test_loss:0.516, Lr:1.00E-04
==================== Done ====================
Epoch:19, Train_acc:78.2%, Train_loss:0.448, Test_acc:76.7%, Test_loss:0.510, Lr:1.00E-04
==================== Done ====================
Epoch:20, Train_acc:78.2%, Train_loss:0.444, Test_acc:76.7%, Test_loss:0.505, Lr:1.00E-04
==================== Done ====================
Epoch:21, Train_acc:78.6%, Train_loss:0.440, Test_acc:77.7%, Test_loss:0.500, Lr:1.00E-04
==================== Done ====================
Epoch:22, Train_acc:78.6%, Train_loss:0.436, Test_acc:78.2%, Test_loss:0.496, Lr:1.00E-04
==================== Done ====================
Epoch:23, Train_acc:78.6%, Train_loss:0.433, Test_acc:78.7%, Test_loss:0.492, Lr:1.00E-04
==================== Done ====================
Epoch:24, Train_acc:79.1%, Train_loss:0.430, Test_acc:79.2%, Test_loss:0.488, Lr:1.00E-04
==================== Done ====================
Epoch:25, Train_acc:79.2%, Train_loss:0.428, Test_acc:79.2%, Test_loss:0.484, Lr:1.00E-04
==================== Done ====================
Epoch:26, Train_acc:79.2%, Train_loss:0.425, Test_acc:79.7%, Test_loss:0.480, Lr:1.00E-04
==================== Done ====================
Epoch:27, Train_acc:79.4%, Train_loss:0.423, Test_acc:79.7%, Test_loss:0.477, Lr:1.00E-04
==================== Done ====================
Epoch:28, Train_acc:79.5%, Train_loss:0.421, Test_acc:79.7%, Test_loss:0.474, Lr:1.00E-04
==================== Done ====================
Epoch:29, Train_acc:79.7%, Train_loss:0.418, Test_acc:78.7%, Test_loss:0.471, Lr:1.00E-04
==================== Done ====================
Epoch:30, Train_acc:79.9%, Train_loss:0.417, Test_acc:78.7%, Test_loss:0.468, Lr:1.00E-04
==================== Done ====================
9.模型评估
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, 5))
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()
扫一扫
6
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
