python
复制代码
import torch
import torch.nn as nn
import torch.optim as optim
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
import time
import matplotlib.pyplot as plt
from tqdm import tqdm
import pandas as pd
data = pd.read_csv(r'data.csv')
list_discrete = data.select_dtypes(include=['object']).columns.tolist()
home_ownership_mapping = {'Own Home': 1, 'Rent': 2,
'Have Mortgage': 3, 'Home Mortgage': 4}
data['Home Ownership'] = data['Home Ownership'].map(home_ownership_mapping)
years_in_job_mapping = {'< 1 year': 1, '1 year': 2, '2 years': 3, '3 years': 4, '4 years': 5,
'5 years': 6, '6 years': 7, '7 years': 8, '8 years': 9, '9 years': 10, '10+ years': 11}
data['Years in current job'] = data['Years in current job'].map(
years_in_job_mapping)
data = pd.get_dummies(data, columns=['Purpose'])
data2 = pd.read_csv(r'data.csv')
list_new = []
for i in data.columns:
if i not in data2.columns:
list_new.append(i)
for i in list_new:
data[i] = data[i].astype(int)
term_mapping = {'Short Term': 0, 'Long Term': 1}
data['Term'] = data['Term'].map(term_mapping)
data.rename(columns={'Term': 'Long Term'}, inplace=True)
list_continuous = data.select_dtypes(
include=['int64', 'float64']).columns.tolist()
for i in list_continuous:
median_value = data[i].median()
data[i] = data[i].fillna(median_value)
X = data.drop(['Credit Default'], axis=1)
y = data['Credit Default']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
print(f'使用设备: {device}\n')
scaler = MinMaxScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)
X_train = torch.FloatTensor(X_train).to(device)
y_train = torch.LongTensor(y_train.values).to(device)
X_test = torch.FloatTensor(X_test).to(device)
y_test = torch.LongTensor(y_test.values).to(device)
class MLP_Original(nn.Module):
def __init__(self):
super(MLP_Original, self).__init__()
self.fc1 = nn.Linear(31, 10)
self.relu = nn.ReLU()
self.fc2 = nn.Linear(10, 3)
def forward(self, x):
out = self.fc1(x)
out = self.relu(out)
out = self.fc2(out)
return out
class MLP_Larger(nn.Module):
def __init__(self):
super(MLP_Larger, self).__init__()
self.fc1 = nn.Linear(31, 20)
self.relu = nn.ReLU()
self.fc2 = nn.Linear(20, 10)
self.fc3 = nn.Linear(10, 3)
def forward(self, x):
out = self.fc1(x)
out = self.relu(out)
out = self.fc2(out)
out = self.relu(out)
out = self.fc3(out)
return out
class MLP_Smaller(nn.Module):
def __init__(self):
super(MLP_Smaller, self).__init__()
self.fc1 = nn.Linear(31, 5)
self.relu = nn.ReLU()
self.fc2 = nn.Linear(5, 3)
def forward(self, x):
out = self.fc1(x)
out = self.relu(out)
out = self.fc2(out)
return out
class MLP_Tanh(nn.Module):
def __init__(self):
super(MLP_Tanh, self).__init__()
self.fc1 = nn.Linear(31, 10)
self.act = nn.Tanh()
self.fc2 = nn.Linear(10, 3)
def forward(self, x):
out = self.fc1(x)
out = self.act(out)
out = self.fc2(out)
return out
def train_and_evaluate(model_class, optimizer_class, lr, num_epochs=20000):
model = model_class().to(device)
criterion = nn.CrossEntropyLoss()
optimizer = optimizer_class(model.parameters(), lr=lr)
losses = []
epochs = []
start_time = time.time()
with tqdm(total=num_epochs, desc=f'训练 {model_class.__name__}', unit='epoch') as pbar:
for epoch in range(num_epochs):
outputs = model(X_train)
loss = criterion(outputs, y_train)
optimizer.zero_grad()
loss.backward()
optimizer.step()
if (epoch + 1) % 200 == 0:
losses.append(loss.item())
epochs.append(epoch + 1)
pbar.set_postfix({'Loss': f'{loss.item():.4f}'})
if (epoch + 1) % 1000 == 0:
pbar.update(1000)
if pbar.n < num_epochs:
pbar.update(num_epochs - pbar.n)
time_all = time.time() - start_time
with torch.no_grad():
outputs = model(X_test)
_, predicted = torch.max(outputs.data, 1)
accuracy = (predicted == y_test).sum().item() / y_test.size(0)
print(f'{model_class.__name__} 训练时间: {time_all:.2f}秒, 测试准确率: {accuracy:.4f}\n')
return epochs, losses, accuracy
configs = [
(MLP_Original, optim.SGD, 0.01),
(MLP_Larger, optim.SGD, 0.01),
(MLP_Smaller, optim.SGD, 0.01),
(MLP_Tanh, optim.SGD, 0.01),
(MLP_Original, optim.Adam, 0.001),
(MLP_Original, optim.SGD, 0.1),
(MLP_Original, optim.SGD, 0.001)
]
plt.figure(figsize=(12, 8))
for config in configs:
epochs, losses, accuracy = train_and_evaluate(*config)
plt.plot(epochs, losses, label=f'{config[0].__name__} {config[1].__name__} lr={config[2]} (Acc:{accuracy:.2f})')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.title('Training Loss Comparison with Different Hyperparameters')
plt.legend()
plt.grid(True)
plt.show()