第 7 章 演習解答¶

import seaborn as sns
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# 1. データの読み込みと先頭5行の表示
tips = sns.load_dataset("tips")
print(tips.head())

# 2. 特徴量 X とラベル y の設定
X = tips[["total_bill", "tip", "size"]]
y = tips["time"]

# 3. 訓練データとテストデータに分割
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

# 4. モデルの作成と学習
model = LogisticRegression(max_iter=1000)
model.fit(X_train, y_train)

# 5. テストデータで予測し、正解率を表示
y_pred = model.predict(X_test)
print(f"正解率: {accuracy_score(y_test, y_pred):.4f}")

# 6. 係数と切片の表示
for name, coef in zip(X.columns, model.coef_[0]):
    print(f"{name}: {coef:.4f}")
print(f"切片: {model.intercept_[0]:.4f}")

# 7. tip の1列だけを特徴量にした場合との比較
X1 = tips[["tip"]]
X1_train, X1_test, y1_train, y1_test = train_test_split(
    X1, y, test_size=0.2, random_state=42
)
model1 = LogisticRegression(max_iter=1000)
model1.fit(X1_train, y1_train)
y1_pred = model1.predict(X1_test)
print(f"正解率（tip のみ）: {accuracy_score(y1_test, y1_pred):.4f}")

import seaborn as sns
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, classification_report

# 1. データの読み込みと特徴量 X・ラベル y の設定
iris = sns.load_dataset("iris")
X = iris[["sepal_length", "sepal_width",
          "petal_length", "petal_width"]]
y = iris["species"]

# 2. 訓練データとテストデータに分割
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

# 3. StandardScaler で標準化
scaler = StandardScaler()
X_train_s = scaler.fit_transform(X_train)
X_test_s = scaler.transform(X_test)

# 4. k-NN モデル（n_neighbors=5）の学習
model = KNeighborsClassifier(n_neighbors=5)
model.fit(X_train_s, y_train)

# 5. テストデータで予測し、正解率を表示
y_pred = model.predict(X_test_s)
print(f"正解率: {accuracy_score(y_test, y_pred):.4f}")

# 6. classification_report の表示
print(classification_report(y_test, y_pred))

# 7. k を変えて訓練・テスト正解率を比較
for k in [1, 3, 5, 7, 9, 11, 15, 21, 31]:
    m = KNeighborsClassifier(n_neighbors=k)
    m.fit(X_train_s, y_train)
    train_acc = accuracy_score(y_train, m.predict(X_train_s))
    test_acc = accuracy_score(y_test, m.predict(X_test_s))
    print(f"k={k:2d}: 訓練 {train_acc:.4f}, テスト {test_acc:.4f}")

import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import (confusion_matrix, precision_score,
                             recall_score, f1_score, classification_report)

# 1. 演習 7-1 と同じモデルを学習し、テストデータを予測
tips = sns.load_dataset("tips")
X = tips[["total_bill", "tip", "size"]]
y = tips["time"]
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)
model = LogisticRegression(max_iter=1000)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

# 2. confusion_matrix の計算と表示
cm = confusion_matrix(y_test, y_pred, labels=["Dinner", "Lunch"])
print("混同行列:")
print(cm)

# 3. ヒートマップで混同行列を可視化
plt.figure(figsize=(5, 4))
sns.heatmap(cm, annot=True, fmt="d", cmap="Blues",
            xticklabels=["Dinner", "Lunch"],
            yticklabels=["Dinner", "Lunch"])
plt.xlabel("Predicted")
plt.ylabel("Actual")
plt.show()

# 4. Lunch を Positive とした適合率・再現率・F1 スコア
precision = precision_score(y_test, y_pred, pos_label="Lunch")
recall = recall_score(y_test, y_pred, pos_label="Lunch")
f1 = f1_score(y_test, y_pred, pos_label="Lunch")
print(f"適合率: {precision:.4f}")
print(f"再現率: {recall:.4f}")
print(f"F1スコア: {f1:.4f}")

# 5. classification_report の表示
print(classification_report(y_test, y_pred))

import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.naive_bayes import GaussianNB
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import accuracy_score, f1_score
import seaborn as sns

# 1. 全特徴量を使用し、標準化を行う
iris = sns.load_dataset("iris")
X = iris[["sepal_length", "sepal_width",
          "petal_length", "petal_width"]].values
y = iris["species"].values

# 2. 訓練データとテストデータに分割
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)
scaler = StandardScaler()
X_train_s = scaler.fit_transform(X_train)
X_test_s = scaler.transform(X_test)

# 3. 4つのモデルの正解率と F1スコアを求める
models = {
    "ロジスティック回帰": LogisticRegression(
        max_iter=200, random_state=42),
    "ナイーブベイズ": GaussianNB(),
    "k-NN (k=5)": KNeighborsClassifier(n_neighbors=5),
    "SVM (RBF)": SVC(kernel="rbf", random_state=42),
}
results = []
for name, model in models.items():
    model.fit(X_train_s, y_train)
    y_pred = model.predict(X_test_s)
    acc = accuracy_score(y_test, y_pred)
    f1 = f1_score(y_test, y_pred, average="macro")
    results.append({
        "モデル": name,
        "正解率": f"{acc:.2%}",
        "F1スコア": f"{f1:.4f}"
    })

# 4. 結果を表にまとめて表示
result_df = pd.DataFrame(results)
print(result_df.to_string(index=False))

import seaborn as sns
from sklearn.svm import SVC
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# 1. データの読み込みと欠損値の削除
penguins = sns.load_dataset("penguins").dropna()

# 2. 特徴量 X とラベル y の設定 + 訓練・テストに分割
X = penguins[["bill_length_mm", "bill_depth_mm", "body_mass_g"]]
y = penguins["species"]
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

# 3. StandardScaler で標準化
scaler = StandardScaler()
X_train_s = scaler.fit_transform(X_train)
X_test_s = scaler.transform(X_test)

# 4. 複数の C で SVM（RBF）を学習し、精度を比較
for C in [0.01, 0.1, 1, 10, 100]:
    model = SVC(kernel="rbf", gamma="scale", C=C, random_state=42)
    model.fit(X_train_s, y_train)
    y_pred = model.predict(X_test_s)
    acc = accuracy_score(y_test, y_pred)
    print(f"C={C:>6}: 正解率 {acc:.4f}")