머신 러닝 (6) 선형 회귀

 import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

import koreanize_matplotlib

import seaborn as sns

from sklearn.datasets import load_diabetes

from sklearn.model_selection import train_test_split

from sklearn.linear_model import LinearRegression

from sklearn.metrics import mean_squared_error

diabetes = load_diabetes()

X = diabetes.data #Feature Matrix

y = diabetes.target #Target Vector

feature_names = diabetes.feature_names

print(type(X))

print(X)

print(X.shape)

df = pd.DataFrame(X, columns=feature_names)

df['target'] = y

display(df.head())

display(df.info())

plt.figure(figsize=(8, 5))

# 타겟 변수의 분포 확인

plt.subplot(1, 2, 1)

sns.histplot(y, kde=True, color='skyblue')

plt.title("원본 Target Distribution")

plt.xlabel("당뇨병 진행도")

# (참고) 로그 변환 적용 시 분포 변화 확인

y_log = np.log1p(y)

plt.subplot(1, 2, 2)

sns.histplot(y_log, kde=True, color='salmon')

plt.title("Log Transformed Target Distribution")

plt.xlabel("Log(Disease Progression)")

plt.tight_layout()

plt.show()

X_train, X_test, y_train, y_test = train_test_split(

    X, y, test_size=0.2, random_state=42

)

X_train.shape, X_test.shape, y_train.shape, y_test.shape

lr = LinearRegression()

lr.fit(X_train, y_train)

y_preds = lr.predict(X_test)

mse = mean_squared_error(y_test, y_preds)

rmse = np.sqrt(mse)

print(f"MSE (Mean Squared Error): {mse:.2f}")

print(f"RMSE (Root Mean Squared Error): {rmse:.2f}")

# (시각화) 실제값 vs 예측값 비교

plt.figure(figsize=(6, 4))

plt.scatter(y_test, y_preds, alpha=0.7, color='green')

plt.plot([y.min(), y.max()], [y.min(), y.max()], 'r--', lw=2) # 대각선 (완벽한 예측선)

plt.xlabel("Actual")

plt.ylabel("Predicted")

plt.title("Actual vs Predicted (Linear Regression)")

plt.show()

print(f"절편 (Intercept): {np.round(lr.intercept_, 2)}")

# 회귀 계수를 Series로 만들어 정렬하여 출력

coef_series = pd.Series(data=np.round(lr.coef_, 2), index=feature_names)

coef_series = coef_series.sort_values(ascending=False)

print("\n[Feature별 회귀 계수 (영향력 순)]")

print(coef_series)

print("\n[해석 예시]")

top_feature = coef_series.index[0]

top_coef = coef_series.iloc[0]

print(f"- 가장 양의 영향력이 큰 변수: {top_feature} (계수: {top_coef})")

print(f"  -> {top_feature} 수치가 높을수록 당뇨병 진행도가 증가하는 경향이 있음.")