クロスバリデーション（交差検証）についてわかりやすく解説

import numpy as np
from sklearn.model_selection import KFold
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error, r2_score

# サンプルデータ生成
np.random.seed(42)
X = np.random.rand(100, 3)
y = 2 * X[:, 0] + 3 * X[:, 1] + 4 * X[:, 2] + np.random.randn(100)

# 5分割クロスバリデーション設定
kf = KFold(n_splits=5, shuffle=True, random_state=42)
model = LinearRegression()

# 各Foldごとのスコアを保存するリスト
mse_list = []
r2_list = []
coeffs_per_fold = []

# クロスバリデーション実行
for fold, (train_idx, test_idx) in enumerate(kf.split(X), start=1):
    X_train, X_test = X[train_idx], X[test_idx]
    y_train, y_test = y[train_idx], y[test_idx]
    
    model.fit(X_train, y_train)
    y_pred = model.predict(X_test)
    
    mse = mean_squared_error(y_test, y_pred)
    r2 = r2_score(y_test, y_pred)
    
    mse_list.append(mse)
    r2_list.append(r2)
    coeffs_per_fold.append(model.coef_)
    
    print(f"Fold {fold}: MSE = {mse:.3f}, R^2 = {r2:.3f}, Coefficients = {model.coef_}")

# 平均スコア出力
average_mse = np.mean(mse_list)
average_r2 = np.mean(r2_list)
average_coeffs = np.mean(coeffs_per_fold, axis=0)

print("\n==== 平均評価指標 ====")
print(f"Average MSE = {average_mse:.3f}")
print(f"Average R^2 = {average_r2:.3f}")
print(f"Average Coefficients = {average_coeffs}")

import numpy as np
from sklearn.model_selection import LeaveOneOut
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error, r2_score

# データ生成
np.random.seed(42)
X = np.random.rand(100, 3)
y = 2 * X[:, 0] + 3 * X[:, 1] + 4 * X[:, 2] + np.random.randn(100)

# モデルとLOOCVの設定
model = LinearRegression()
loo = LeaveOneOut()

# 予測結果と係数の格納用
y_pred = np.zeros_like(y)
coefs = np.zeros((len(y), X.shape[1]))  # 100 × 3

# LOOCVの実行
for i, (train_idx, test_idx) in enumerate(loo.split(X)):
    X_train, X_test = X[train_idx], X[test_idx]
    y_train = y[train_idx]
    
    model.fit(X_train, y_train)
    y_pred[test_idx] = model.predict(X_test)
    coefs[i] = model.coef_

# MSEとR²の計算
mse = mean_squared_error(y, y_pred)
r2 = r2_score(y, y_pred)

# 係数の平均計算
mean_coefs = np.mean(coefs, axis=0)

# 結果の出力
print(f"LOOCV - MSE: {mse:.3f}")
print(f"LOOCV - R²: {r2:.3f}")
print(f"LOOCV - 平均係数: {mean_coefs}")

import numpy as np
from sklearn.datasets import make_classification
from sklearn.model_selection import StratifiedKFold
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

# 仮想のデータ生成
X, y = make_classification(n_samples=1000, n_features=20, n_classes=2, random_state=42)

# 層化k分割交差検証を設定
num_folds = 5
skf = StratifiedKFold(n_splits=num_folds, shuffle=True, random_state=42)

# ロジスティック回帰モデルを設定
model = LogisticRegression()

# 各フォールドでの性能評価を実行
fold_accuracy_scores = []
for fold, (train_index, test_index) in enumerate(skf.split(X, y), start=1):
    X_train, X_test = X[train_index], X[test_index]
    y_train, y_test = y[train_index], y[test_index]
    
    model.fit(X_train, y_train)
    y_pred = model.predict(X_test)
    accuracy = accuracy_score(y_test, y_pred)
    fold_accuracy_scores.append(accuracy)
    
    print(f"Fold {fold}: Accuracy = {accuracy}")

# 平均精度を計算して表示
average_accuracy = np.mean(fold_accuracy_scores)
print(f"Average Accuracy across {num_folds} folds = {average_accuracy}")

'''
Fold 1: Accuracy = 0.87
Fold 2: Accuracy = 0.885
Fold 3: Accuracy = 0.87
Fold 4: Accuracy = 0.86
Fold 5: Accuracy = 0.86
Average Accuracy across 5 folds = 0.869
'''