Gaussian Mixture

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79

from __future__ import annotations

import japanize_matplotlib
import matplotlib.pyplot as plt
import numpy as np
from numpy.typing import NDArray
from sklearn.datasets import make_blobs
from sklearn.mixture import GaussianMixture

def run_gmm_demo(
    n_samples: int = 600,
    n_components: int = 3,
    cluster_std: list[float] | tuple[float, ...] = (1.0, 1.4, 0.8),
    covariance_type: str = "full",
    random_state: int = 7,
    n_init: int = 8,
) -> dict[str, object]:
    """เรียนรู้ Gaussian Mixture และแสดงศูนย์กลางกับหน้าที่รับผิดชอบ."""
    japanize_matplotlib.japanize()
    features, labels_true = make_blobs(
        n_samples=n_samples,
        centers=n_components,
        cluster_std=cluster_std,
        random_state=random_state,
    )

    gmm = GaussianMixture(
        n_components=n_components,
        covariance_type=covariance_type,
        random_state=random_state,
        n_init=n_init,
    )
    gmm.fit(features)

    hard_labels = gmm.predict(features)
    responsibilities = gmm.predict_proba(features)
    log_likelihood = float(gmm.score(features))
    weights = gmm.weights_

    fig, ax = plt.subplots(figsize=(6.2, 5.2))
    scatter = ax.scatter(
        features[:, 0],
        features[:, 1],
        c=hard_labels,
        cmap="viridis",
        s=30,
        edgecolor="white",
        linewidth=0.2,
        alpha=0.85,
    )
    ax.scatter(
        gmm.means_[:, 0],
        gmm.means_[:, 1],
        marker="x",
        c="red",
        s=140,
        linewidth=2.0,
        label="ศูนย์กลางคลัสเตอร์",
    )
    ax.set_title("Soft clustering ด้วย Gaussian Mixture")
    ax.set_xlabel("คุณลักษณะที่ 1")
    ax.set_ylabel("คุณลักษณะที่ 2")
    ax.grid(alpha=0.2)
    handles, _ = scatter.legend_elements()
    labels = [f"คลัสเตอร์ {idx}" for idx in range(n_components)]
    ax.legend(handles, labels, title="ป้ายที่พยากรณ์", loc="upper right")
    fig.tight_layout()
    plt.show()

    return {
        "log_likelihood": log_likelihood,
        "weights": weights.tolist(),
        "responsibilities_shape": responsibilities.shape,
    }

metrics = run_gmm_demo()
print(f"ค่า log-likelihood: {metrics['log_likelihood']:.3f}")
print("ค่าน้ำหนักของส่วนผสม:", metrics["weights"])
print("รูปทรงของเมทริกซ์หน้าที่รับผิดชอบ:", metrics["responsibilities_shape"])