מדעי הנתונים למתחילים: מפת הדרכים המלאה

1import numpy as np
2import matplotlib.pyplot as plt
3from sklearn.datasets import make_moons
4from sklearn.model_selection import train_test_split
5from sklearn.neighbors import KNeighborsClassifier
6

7X, y = make_moons(n_samples=200, noise=0.3, random_state=0)
8X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)
9

10k = 5
11model = KNeighborsClassifier(n_neighbors=k)
12model.fit(X_train, y_train)
13print(f"train accuracy: {model.score(X_train, y_train):.2f}")
14print(f"test accuracy:  {model.score(X_test, y_test):.2f}")
15

16xx, yy = np.meshgrid(np.linspace(-2, 3, 56), np.linspace(-1.5, 2, 40))
17proba = model.predict_proba(np.c_[xx.ravel(), yy.ravel()])[:, 1]
18proba = proba.reshape(xx.shape)
19

20plt.figure(figsize=(5, 4))
21plt.pcolormesh(xx, yy, proba, cmap="coolwarm", alpha=0.35)
22plt.scatter(X_train[:, 0], X_train[:, 1], c=y_train,
23            cmap="coolwarm", edgecolors="k")
24plt.title(f"k-nearest neighbors, k = {k}")
25plt.show()