การถดถอยเชิงเส้นและวิธีกำลังสองน้อยที่สุด | พื้นฐานทฤษฎีและโค้ด Python

from __future__ import annotations

import japanize_matplotlib
import matplotlib.pyplot as plt
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler


def plot_simple_linear_regression(n_samples: int = 100) -> None:
    """Plot a fitted linear regression model for synthetic data.

    Args:
        n_samples: Number of synthetic samples to generate.
    """
    japanize_matplotlib.japanize()
    rng = np.random.default_rng(seed=0)

    X: np.ndarray = np.linspace(-5.0, 5.0, n_samples, dtype=float)[:, np.newaxis]
    noise: np.ndarray = rng.normal(scale=2.0, size=n_samples)
    y: np.ndarray = 2.0 * X.ravel() + 1.0 + noise

    model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())
    model.fit(X, y)
    y_pred: np.ndarray = model.predict(X)

    fig, ax = plt.subplots(figsize=(10, 5))
    ax.scatter(X, y, marker="x", label="ข้อมูลจริง", c="orange")
    ax.plot(X, y_pred, label="ผลพยากรณ์เชิงเส้น")
    ax.set_xlabel("$x$")
    ax.set_ylabel("$y$")
    ax.legend()
    fig.tight_layout()
    plt.show()


plot_simple_linear_regression()