第R3周：RNN-心脏病预测

• 🍨 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、前期准备工作

1. 设置硬件设备

import numpy as np
import pandas as pd
import torch
from torch import nn
import torch.nn.functional as F
import seaborn as sns

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device

device(type=‘cpu’)

2. 导入数据

df = pd.read_csv("heart.csv")

df

303 rows × 14 columns

二、 构建数据集

1. 标准化

from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split

X = df.iloc[:,: -1]
y = df.iloc[:, -1]

# Standardize the data
sc = StandardScaler()
X = sc.fit_transform(X)

2. 划分数据集

X = torch.tensor(np.array(X), dtype=torch.float32)
y = torch.tensor(np.array(y), dtype=torch.int64)

X_train, X_test, y_train, y_test = train_test_split(X, y, 
                                                    test_size=0.1,
                                                    random_state=1)

X_train.shape, y_train.shape

(torch.Size([272, 13]), torch.Size([272]))

3. 构建数据加载器

from torch.utils.data import TensorDataset, DataLoader

train_dl = DataLoader(TensorDataset(X_train, y_train),
                      batch_size = 64, 
                      shuffle = False)

test_dl = DataLoader(TensorDataset(X_test, y_test),
                     batch_size = 64,
                     shuffle = False)

三、 模型训练

1. 构建模型

class model_rnn(nn.Module):
    def __init__(self):
        super(model_rnn, self).__init__()
        self.rnn0 = nn.RNN(input_size=13, hidden_size=200,
                           num_layers=1, batch_first=True)

        self.fc0 = nn.Linear(200, 50)
        self.fc1 = nn.Linear(50, 2)

    def forward(self, x):

        out, hidden1 = self.rnn0(x)
        out = self.fc0(out)
        out = self.fc1(out)

        return out

model = model_rnn().to(device)
model

model_rnn(
(rnn0): RNN(13, 200, batch_first=True)
(fc0): Linear(in_features=200, out_features=50, bias=True)
(fc1): Linear(in_features=50, out_features=2, bias=True)
)

model(torch.rand(30,13).to(device)).shape

torch.Size([30, 2])

2. 定义训练函数

# 训练循环
def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)
    num_batches = len(dataloader)

    train_loss, train_acc = 0, 0

    for X, y in dataloader:
        X, y = X.to(device), y.to(device)

        pred = model(X)
        loss = loss_fn(pred, y)

        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        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

3.定义测试函数

def test (dataloader, model, loss_fn):
    size = len(dataloader.dataset)
    num_batches = len(dataloader)
    test_loss, test_acc = 0, 0

    with torch.no_grad():
        for imgs, target in dataloader:
            imgs, target = imgs.to(device), target.to(device)

            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
            

4. 正式训练模型

loss_fn = nn.CrossEntropyLoss()
learn_rate = 1e-4
opt = torch.optim.Adam(model.parameters(), lr=learn_rate)
epochs = 50

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)

    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:41.9%, Train_loss:0.699, Test_acc:61.3%, Test_loss:0.679, Lr:1.00E-04
==================== Done ====================
Epoch: 2, Train_acc:48.9%, Train_loss:0.685, Test_acc:67.7%, Test_loss:0.664, Lr:1.00E-04
==================== Done ====================
Epoch: 3, Train_acc:59.6%, Train_loss:0.673, Test_acc:67.7%, Test_loss:0.649, Lr:1.00E-04
==================== Done ====================
Epoch: 4, Train_acc:69.5%, Train_loss:0.661, Test_acc:83.9%, Test_loss:0.635, Lr:1.00E-04
==================== Done ====================
Epoch: 5, Train_acc:74.6%, Train_loss:0.649, Test_acc:83.9%, Test_loss:0.622, Lr:1.00E-04
==================== Done ====================
Epoch: 6, Train_acc:77.6%, Train_loss:0.637, Test_acc:87.1%, Test_loss:0.608, Lr:1.00E-04
==================== Done ====================
Epoch: 7, Train_acc:80.1%, Train_loss:0.625, Test_acc:87.1%, Test_loss:0.595, Lr:1.00E-04
==================== Done ====================
Epoch: 8, Train_acc:81.6%, Train_loss:0.613, Test_acc:83.9%, Test_loss:0.581, Lr:1.00E-04
==================== Done ====================
Epoch: 9, Train_acc:81.6%, Train_loss:0.600, Test_acc:83.9%, Test_loss:0.568, Lr:1.00E-04
==================== Done ====================
Epoch:10, Train_acc:81.6%, Train_loss:0.588, Test_acc:83.9%, Test_loss:0.555, Lr:1.00E-04
==================== Done ====================
Epoch:11, Train_acc:82.4%, Train_loss:0.575, Test_acc:83.9%, Test_loss:0.542, Lr:1.00E-04
==================== Done ====================
Epoch:12, Train_acc:83.1%, Train_loss:0.561, Test_acc:83.9%, Test_loss:0.529, Lr:1.00E-04
==================== Done ====================
Epoch:13, Train_acc:83.5%, Train_loss:0.548, Test_acc:83.9%, Test_loss:0.517, Lr:1.00E-04
==================== Done ====================
Epoch:14, Train_acc:83.8%, Train_loss:0.534, Test_acc:83.9%, Test_loss:0.506, Lr:1.00E-04
==================== Done ====================
Epoch:15, Train_acc:83.5%, Train_loss:0.520, Test_acc:83.9%, Test_loss:0.495, Lr:1.00E-04
==================== Done ====================
Epoch:16, Train_acc:83.5%, Train_loss:0.505, Test_acc:83.9%, Test_loss:0.486, Lr:1.00E-04
==================== Done ====================
Epoch:17, Train_acc:83.5%, Train_loss:0.491, Test_acc:83.9%, Test_loss:0.478, Lr:1.00E-04
==================== Done ====================
Epoch:18, Train_acc:83.8%, Train_loss:0.477, Test_acc:83.9%, Test_loss:0.471, Lr:1.00E-04
==================== Done ====================
Epoch:19, Train_acc:83.5%, Train_loss:0.463, Test_acc:83.9%, Test_loss:0.465, Lr:1.00E-04
==================== Done ====================
Epoch:20, Train_acc:83.1%, Train_loss:0.450, Test_acc:80.6%, Test_loss:0.459, Lr:1.00E-04
==================== Done ====================
Epoch:21, Train_acc:83.5%, Train_loss:0.437, Test_acc:80.6%, Test_loss:0.454, Lr:1.00E-04
==================== Done ====================
Epoch:22, Train_acc:83.8%, Train_loss:0.424, Test_acc:80.6%, Test_loss:0.449, Lr:1.00E-04
==================== Done ====================
Epoch:23, Train_acc:84.2%, Train_loss:0.412, Test_acc:80.6%, Test_loss:0.442, Lr:1.00E-04
==================== Done ====================
Epoch:24, Train_acc:84.2%, Train_loss:0.399, Test_acc:83.9%, Test_loss:0.434, Lr:1.00E-04
==================== Done ====================
Epoch:25, Train_acc:84.2%, Train_loss:0.387, Test_acc:87.1%, Test_loss:0.426, Lr:1.00E-04
==================== Done ====================
Epoch:26, Train_acc:84.6%, Train_loss:0.376, Test_acc:87.1%, Test_loss:0.419, Lr:1.00E-04
==================== Done ====================
Epoch:27, Train_acc:85.3%, Train_loss:0.365, Test_acc:87.1%, Test_loss:0.412, Lr:1.00E-04
==================== Done ====================
Epoch:28, Train_acc:86.0%, Train_loss:0.354, Test_acc:87.1%, Test_loss:0.406, Lr:1.00E-04
==================== Done ====================
Epoch:29, Train_acc:85.3%, Train_loss:0.343, Test_acc:87.1%, Test_loss:0.401, Lr:1.00E-04
==================== Done ====================
Epoch:30, Train_acc:87.1%, Train_loss:0.333, Test_acc:87.1%, Test_loss:0.397, Lr:1.00E-04
==================== Done ====================
Epoch:31, Train_acc:87.5%, Train_loss:0.324, Test_acc:87.1%, Test_loss:0.394, Lr:1.00E-04
==================== Done ====================
Epoch:32, Train_acc:88.2%, Train_loss:0.314, Test_acc:87.1%, Test_loss:0.390, Lr:1.00E-04
==================== Done ====================
Epoch:33, Train_acc:88.2%, Train_loss:0.306, Test_acc:87.1%, Test_loss:0.388, Lr:1.00E-04
==================== Done ====================
Epoch:34, Train_acc:88.6%, Train_loss:0.297, Test_acc:87.1%, Test_loss:0.386, Lr:1.00E-04
==================== Done ====================
Epoch:35, Train_acc:89.0%, Train_loss:0.289, Test_acc:87.1%, Test_loss:0.384, Lr:1.00E-04
==================== Done ====================
Epoch:36, Train_acc:88.6%, Train_loss:0.282, Test_acc:87.1%, Test_loss:0.384, Lr:1.00E-04
==================== Done ====================
Epoch:37, Train_acc:89.0%, Train_loss:0.274, Test_acc:83.9%, Test_loss:0.384, Lr:1.00E-04
==================== Done ====================
Epoch:38, Train_acc:89.3%, Train_loss:0.267, Test_acc:80.6%, Test_loss:0.385, Lr:1.00E-04
==================== Done ====================
Epoch:39, Train_acc:89.3%, Train_loss:0.261, Test_acc:80.6%, Test_loss:0.386, Lr:1.00E-04
==================== Done ====================
Epoch:40, Train_acc:90.1%, Train_loss:0.254, Test_acc:80.6%, Test_loss:0.389, Lr:1.00E-04
==================== Done ====================
Epoch:41, Train_acc:90.1%, Train_loss:0.247, Test_acc:80.6%, Test_loss:0.392, Lr:1.00E-04
==================== Done ====================
Epoch:42, Train_acc:90.1%, Train_loss:0.240, Test_acc:80.6%, Test_loss:0.396, Lr:1.00E-04
==================== Done ====================
Epoch:43, Train_acc:90.1%, Train_loss:0.234, Test_acc:77.4%, Test_loss:0.400, Lr:1.00E-04
==================== Done ====================
Epoch:44, Train_acc:90.4%, Train_loss:0.227, Test_acc:77.4%, Test_loss:0.404, Lr:1.00E-04
==================== Done ====================
Epoch:45, Train_acc:91.9%, Train_loss:0.221, Test_acc:77.4%, Test_loss:0.409, Lr:1.00E-04
==================== Done ====================
Epoch:46, Train_acc:93.4%, Train_loss:0.214, Test_acc:77.4%, Test_loss:0.414, Lr:1.00E-04
==================== Done ====================
Epoch:47, Train_acc:93.4%, Train_loss:0.208, Test_acc:77.4%, Test_loss:0.418, Lr:1.00E-04
==================== Done ====================
Epoch:48, Train_acc:93.8%, Train_loss:0.202, Test_acc:77.4%, Test_loss:0.422, Lr:1.00E-04
==================== Done ====================
Epoch:49, Train_acc:93.8%, Train_loss:0.197, Test_acc:77.4%, Test_loss:0.426, Lr:1.00E-04
==================== Done ====================
Epoch:50, Train_acc:94.5%, Train_loss:0.191, Test_acc:77.4%, Test_loss:0.429, Lr:1.00E-04
==================== Done ====================

四、模型评估

1. Loss与Accuracy图

import matplotlib.pyplot as plt
from datetime import datetime
#隐藏警告
import warnings
warnings.filterwarnings("ignore")        #忽略警告信息

current_time = datetime.now() # 获取当前时间

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']         = 200        #分辨率

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()

2. 混淆矩阵

print("==============输入数据Shape为==============")
print("X_test.shape：",X_test.shape)
print("y_test.shape：",y_test.shape)

pred = model(X_test.to(device)).argmax(1).cpu().numpy()

print("\n==============输出数据Shape为==============")
print("pred.shape：",pred.shape)

输入数据Shape为
X_test.shape： torch.Size([31, 13])
y_test.shape： torch.Size([31])

输出数据Shape为
pred.shape： (31,)

import numpy as np
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay

# 计算混淆矩阵
cm = confusion_matrix(y_test, pred)

plt.figure(figsize=(6,5))
plt.suptitle('')
sns.heatmap(cm, annot=True, fmt="d", cmap="Blues")

# 修改字体大小
plt.xticks(fontsize=10)
plt.yticks(fontsize=10)
plt.title("Confusion Matrix", fontsize=12)
plt.xlabel("Predicted Label", fontsize=10)
plt.ylabel("True Label", fontsize=10)

# 显示图
plt.tight_layout()  # 调整布局防止重叠
plt.show()

3. 调用模型进行预测

test_X = X_test[0].reshape(1, -1) # X_test[0]即我们的输入数据
 
pred = model(test_X.to(device)).argmax(1).item()
print("模型预测结果为：",pred)
print("=="*20)
print("0：不会患心脏病")
print("1：可能患心脏病")

模型预测结果为： 0
========================================
0：不会患心脏病
1：可能患心脏病

五、总结

本周主要学习了LSTM和RNN，通过实践项目更加深入地了解了RNN模型的结构。