当前位置：首页 > news >正文

365打卡第R8周: RNN实现阿尔茨海默病诊断

news 来源：原创 2025/5/11 7:40:25

🍨 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

🏡 我的环境：

语言环境：Python3.10
编译器：Jupyter Lab
深度学习环境：torch==2.5.1 torchvision==0.20.1
------------------------------分割线---------------------------------

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import torch
import torch.nn as nn
import torch.nn.functional as F# 设置GPU
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(device)

(数据规模很小，在普通笔记本电脑就可以流畅跑完，不用GPU也行)

df = pd.read_csv('./alzheimers_disease_data.csv')
# 删除最后一列和第一列
df = df.iloc[:, 1:-1]
df.head()

from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScalerX = df.iloc[:, :-1]
y = df.iloc[:, -1]# 将每一列特征标准化为标准正态分布，注意，标准化是针对每一列而言的
scaler = StandardScaler()
X = scaler.fit_transform(X)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

from torch.utils.data import TensorDataset, DataLoadertrain_dl = DataLoader(TensorDataset(X_train, y_train), batch_size=32, shuffle=False)
test_dl = DataLoader(TensorDataset(X_test, y_test), batch_size=32, shuffle=False)

class model_rnn(nn.Module):def __init__(self):super(model_rnn, self).__init__()self.rnn0 = nn.RNN(input_size=32, 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 outmodel = model_rnn().to(device)
model

# 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_lossdef 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

loss_fn    = nn.CrossEntropyLoss() # 创建损失函数
learn_rate = 5e-5
opt = torch.optim.Adam(model.parameters(), lr= learn_rate)epochs     = 50train_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('Done')

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()

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

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('Confusion Matrix')
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues')# 修改字体大小
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.title('Confusion Matrix', fontsize=12)
plt.xlabel('Predicted Label', fontsize=12)
plt.ylabel('True Labels', fontsize=10)# 调整布局防止重叠
plt.tight_layout()# 显示图形
plt.show()

# 调用模型进行预测
test_X = X_test[0].reshape(1, -1)pred = model(test_X.to(device)).argmax(1).item()
print("模型预测结果为： ",pred)
print("="*20)
print("0: 未患病")
print("1: 已患病")

---------------------------小结和优化（顺带完成R9周的内容）--------------------------

1、本周的代码定义了一个RNN模型，结构是RNN层接两个全连接层。使用交叉熵损失和Adam优化器，进行了50个epoch的训练。

2、RNN的输入通常需要三维数据（batch_size, sequence_length, features），但原代码中的是二维的（样本数×特征数），所以在输入RNN的时候通常视为sqn_len=1进行处理，可能并不能充分发挥RNN的能力。

3、尝试结合RFE进行特征选择，然后用选出的特征训练RNN。

--------------------------------

1. 修改数据预处理流程

# 正确划分训练集和测试集后标准化
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=1)scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)  # 仅用训练集的均值和方差

2. 使用RFE选择重要特征

from sklearn.feature_selection import RFE
from sklearn.linear_model import LogisticRegressionn_features_to_select = 10  # 可根据需求调整
selector = RFE(estimator=LogisticRegression(max_iter=1000),n_features_to_select=n_features_to_select,step=1
)
selector.fit(X_train_scaled, y_train)# 筛选特征
X_train_selected = selector.transform(X_train_scaled)
X_test_selected = selector.transform(X_test_scaled)

3. 调整数据维度以适应RNN

# 转换为三维张量 (batch, seq_len=1, features)
X_train_tensor = torch.tensor(X_train_selected, dtype=torch.float32).unsqueeze(1)
X_test_tensor = torch.tensor(X_test_selected, dtype=torch.float32).unsqueeze(1)
y_train_tensor = torch.tensor(y_train.values, dtype=torch.int64)
y_test_tensor = torch.tensor(y_test.values, dtype=torch.int64)

4. 动态调整模型输入尺寸

class model_rnn(nn.Module):def __init__(self, input_size):  # 根据特征数动态调整super(model_rnn, self).__init__()self.rnn0 = nn.RNN(input_size=input_size,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, _ = self.rnn0(x)out = self.fc0(out[:, -1, :])  # 取最后一个时间步输出out = self.fc1(out)return out# 创建模型时传入特征数
input_size = X_train_selected.shape[1]
model = model_rnn(input_size).to(device)

优化后运行结果