机器学习实战：如何准确判断2组数据过拟合问题

1次阅读

没有评论

共计 2640 个字符，预计需要花费 7 分钟才能阅读完成。

在机器学习项目中，过拟合是一个让人头疼的问题。简单来说，过拟合就是模型在训练数据上表现很好，但在新数据上表现糟糕。这种情况在实际开发中经常遇到，尤其是当数据量较小或者模型复杂度较高时。

直接影响 ：过拟合会导致模型在实际应用中失去预测能力，浪费开发资源
判断难点 ：很多开发者难以准确区分模型是真正学习到了规律，还是仅仅记住了训练数据
典型场景 ：当我们只有 2 组数据（训练集和测试集）时，判断过拟合尤为困难

原理：将数据分成 k 份，轮流用 k - 1 份训练，1 份验证
优点：充分利用数据，评估更稳定
缺点：计算成本较高

原理：观察训练集和验证集误差随样本量的变化
优点：直观显示过拟合趋势
缺点：需要多次训练模型

原理：预留固定验证集监控模型性能
优点：实现简单
缺点：数据利用率低

# 导入必要库
from sklearn.model_selection import train_test_split, learning_curve
from sklearn.linear_model import LogisticRegression
from sklearn.datasets import make_classification
import matplotlib.pyplot as plt
import numpy as np

# 生成模拟数据
X, y = make_classification(n_samples=1000, n_features=20, n_informative=2, random_state=42)

# 数据分割
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# 训练模型
model = LogisticRegression(max_iter=1000)
model.fit(X_train, y_train)

# 评估性能
train_score = model.score(X_train, y_train)
test_score = model.score(X_test, y_test)
print(f"训练集准确率: {train_score:.3f}, 测试集准确率: {test_score:.3f}")

# 计算学习曲线
train_sizes, train_scores, test_scores = learning_curve(model, X, y, cv=5, train_sizes=np.linspace(0.1, 1.0, 10))

# 计算均值和标准差
train_mean = np.mean(train_scores, axis=1)
train_std = np.std(train_scores, axis=1)
test_mean = np.mean(test_scores, axis=1)
test_std = np.std(test_scores, axis=1)

# 绘制学习曲线
plt.figure(figsize=(10, 6))
plt.plot(train_sizes, train_mean, color='blue', marker='o', label='训练准确率')
plt.fill_between(train_sizes, train_mean + train_std, train_mean - train_std, alpha=0.15, color='blue')
plt.plot(train_sizes, test_mean, color='green', linestyle='--', marker='s', label='验证准确率')
plt.fill_between(train_sizes, test_mean + test_std, test_mean - test_std, alpha=0.15, color='green')
plt.xlabel('训练样本数')
plt.ylabel('准确率')
plt.legend(loc='lower right')
plt.show()

from sklearn.linear_model import Ridge

# 使用 L2 正则化
ridge = Ridge(alpha=1.0)
ridge.fit(X_train, y_train)

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout

model = Sequential([Dense(64, activation='relu', input_shape=(20,)),
    Dropout(0.5),
    Dense(1, activation='sigmoid')
])

from tensorflow.keras.callbacks import EarlyStopping

early_stopping = EarlyStopping(monitor='val_loss', patience=5)
model.fit(X_train, y_train, validation_data=(X_test, y_test), callbacks=[early_stopping])