8. 木構造とアンサンブル学習¶

import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.tree import DecisionTreeClassifier, plot_tree
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
X = penguins[["bill_length_mm", "bill_depth_mm", "body_mass_g"]]
y = penguins["species"]

# 3. 訓練・テストに分割（8:2, random_state=42）
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

# 4. 決定木モデル（max_depth=3）の学習
model = DecisionTreeClassifier(max_depth=3, random_state=42)
model.fit(X_train, y_train)

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

# 6. plot_tree で木構造を可視化
plt.figure(figsize=(14, 8))
plot_tree(model, feature_names=X.columns, class_names=model.classes_,
          filled=True, fontsize=9)
plt.show()

# 7. max_depth を 1〜10 で変化させ、訓練精度とテスト精度をプロット
depths = range(1, 11)
train_accs, test_accs = [], []
for depth in depths:
    clf = DecisionTreeClassifier(max_depth=depth, random_state=42)
    clf.fit(X_train, y_train)
    train_accs.append(accuracy_score(y_train, clf.predict(X_train)))
    test_accs.append(accuracy_score(y_test, clf.predict(X_test)))
    print(f"max_depth={depth:2d}: 訓練 {train_accs[-1]:.4f}, "
          f"テスト {test_accs[-1]:.4f}")

plt.figure(figsize=(8, 5))
plt.plot(depths, train_accs, marker="o", label="訓練精度")
plt.plot(depths, test_accs, marker="s", label="テスト精度")
plt.xlabel("max_depth")
plt.ylabel("正解率")
plt.legend()
plt.grid(True, alpha=0.3)
plt.show()

import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# 1. データの準備（演習8-2と同じ）
penguins = sns.load_dataset("penguins").dropna()
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
)

# 2. ランダムフォレストの学習と正解率の表示
model = RandomForestClassifier(n_estimators=100, max_depth=3,
                               random_state=42)
model.fit(X_train, y_train)
print(f"正解率: {accuracy_score(y_test, model.predict(X_test)):.4f}")

# 3. 特徴量の重要度を DataFrame に整理して表示
importance_df = pd.DataFrame({
    "特徴量": X.columns,
    "重要度": model.feature_importances_
}).sort_values("重要度", ascending=False)
print(importance_df)

# 4. 棒グラフで可視化
plt.figure(figsize=(8, 4))
plt.barh(importance_df["特徴量"], importance_df["重要度"])
plt.xlabel("重要度")
plt.gca().invert_yaxis()
plt.show()

# 5. max_depth を変えたときのテスト精度を比較
for depth in [1, 2, 3, 5, None]:
    clf = RandomForestClassifier(n_estimators=100, max_depth=depth,
                                 random_state=42)
    clf.fit(X_train, y_train)
    acc = accuracy_score(y_test, clf.predict(X_test))
    print(f"max_depth={str(depth):>4s}: テスト精度 {acc:.4f}")

import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.metrics import accuracy_score

# ペンギンデータセット（演習8-2と同じ準備）
penguins = sns.load_dataset("penguins").dropna()
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
)

for n in [10, 50, 100, 500, 1000]:
    clf = GradientBoostingClassifier(
        n_estimators=n, learning_rate=0.01,
        max_depth=3, random_state=42
    )
    clf.fit(X_train, y_train)
    acc = accuracy_score(y_test, clf.predict(X_test))
    print(f"n_estimators={n:4d}: テスト精度 {acc:.4f}")

import seaborn as sns
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import (RandomForestClassifier,
                              GradientBoostingClassifier)

# irisデータセット
df = sns.load_dataset("iris")
X = df.drop("species", axis=1)
y = df["species"]
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42
)

# 3つのモデルを学習し、特徴量重要度を集める
models = {
    "決定木": DecisionTreeClassifier(max_depth=3, random_state=42),
    "ランダムフォレスト": RandomForestClassifier(
        n_estimators=100, random_state=42),
    "勾配ブースティング": GradientBoostingClassifier(
        n_estimators=100, learning_rate=0.1,
        max_depth=3, random_state=42),
}
importance_df = pd.DataFrame(index=X.columns)
for name, model in models.items():
    model.fit(X_train, y_train)
    importance_df[name] = model.feature_importances_

print(importance_df.round(4))

# 棒グラフで比較
fig, axes = plt.subplots(1, 3, figsize=(15, 4))
for ax, name in zip(axes, models):
    importances = importance_df[name].values
    indices = np.argsort(importances)
    ax.barh(range(len(indices)), importances[indices])
    ax.set_yticks(range(len(indices)))
    ax.set_yticklabels(importance_df.index[indices])
    ax.set_xlabel("重要度")
    ax.set_title(name)
plt.tight_layout()
plt.show()