推荐系统中的正则化技术

import numpy as np
from sklearn.linear_model import Lasso
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error

# 示例数据
X, y = np.random.randn(100, 10), np.random.randn(100)

# 数据划分
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Lasso模型
lasso = Lasso(alpha=0.1)
lasso.fit(X_train, y_train)

# 预测
y_pred = lasso.predict(X_test)

# 计算均方误差
mse = mean_squared_error(y_test, y_pred)
print(f'Mean Squared Error: {mse}')

from sklearn.linear_model import Ridge

# Ridge模型
ridge = Ridge(alpha=0.1)
ridge.fit(X_train, y_train)

# 预测
y_pred = ridge.predict(X_test)

# 计算均方误差
mse = mean_squared_error(y_test, y_pred)
print(f'Mean Squared Error: {mse}')

from sklearn.linear_model import ElasticNet

# ElasticNet模型
elastic_net = ElasticNet(alpha=0.1, l1_ratio=0.5)
elastic_net.fit(X_train, y_train)

# 预测
y_pred = elastic_net.predict(X_test)

# 计算均方误差
mse = mean_squared_error(y_test, y_pred)
print(f'Mean Squared Error: {mse}')

import torch
import torch.nn as nn
import torch.optim as optim

# 简单神经网络模型
class SimpleNN(nn.Module):
    def __init__(self):
        super(SimpleNN, self).__init__()
        self.fc1 = nn.Linear(10, 50)
        self.dropout = nn.Dropout(0.5)
        self.fc2 = nn.Linear(50, 1)

    def forward(self, x):
        x = torch.relu(self.fc1(x))
        x = self.dropout(x)
        x = self.fc2(x)
        return x

# 模型训练
model = SimpleNN()
criterion = nn.MSELoss()
optimizer = optim.SGD(model.parameters(), lr=0.01)

# 示例数据
X_train_torch = torch.tensor(X_train, dtype=torch.float32)
y_train_torch = torch.tensor(y_train, dtype=torch.float32)

# 训练循环
for epoch in range(100):
    optimizer.zero_grad()
    output = model(X_train_torch)
    loss = criterion(output, y_train_torch.unsqueeze(1))
    loss.backward()
    optimizer.step()

# 预测
X_test_torch = torch.tensor(X_test, dtype=torch.float32)
y_pred_torch = model(X_test_torch)
mse_torch = mean_squared_error(y_test, y_pred_torch.detach().numpy())
print(f'Mean Squared Error: {mse_torch}')

import pandas as pd
from sklearn.decomposition import NMF

# 读取数据
ratings = pd.read_csv('ratings.csv')

# 创建用户-物品矩阵
user_item_matrix = ratings.pivot(index='userId', columns='movieId', values='rating').fillna(0)

# 使用NMF进行矩阵分解
nmf_model = NMF(n_components=20, alpha=0.1, l1_ratio=0.5)
user_matrix = nmf_model.fit_transform(user_item_matrix)
item_matrix = nmf_model.components_

# 预测评分
predicted_ratings = np.dot(user_matrix, item_matrix)

# 计算MSE
actual_ratings = user_item_matrix.values
mse_nmf = mean_squared_error(actual_ratings[actual_ratings > 0], predicted_ratings[actual_ratings > 0])
print(f'Mean Squared Error with Regularization: {mse_nmf}')