Statistical Learning Methods Using Python

from scipy import stats
from pylab import *
def knnClassify(Target_feature, dataSet, p, k):    dataSet_x = dataSet[:, 0:-1]    dataSet_y = dataSet[:, -1]    L_p = []    for i in range(len(dataSet_x)):        l_p = 0        for x_i in range(len(dataSet_x[i])):            l_p = (abs(dataSet_x[i][x_i] - Target_feature[x_i]))**p+l_p        l_p =math.pow(l_p, 1/p)        L_p.append(l_p)    L_p = np.array(L_p)    y_i = L_p.argsort()    y = []    max_L_p = []    for k_i in range(0,k):        y.append(dataSet_y[y_i[k_i]])        max_L_p.append(L_p[y_i[k_i]])    print(y)    max_L = max(max_L_p)    y_result = stats.mode(y)[0][0]    return y_result,max_L
if __name__ == '__main__':    Target_feature = np.array([1, 3])    dataSet = np.array([[1, 4, 1], [0.5, 2, 1], [2, 2.3, 1], [1, 0.5, -1], [2, 1, -1],            [4, 1, -1], [3.5, 4, 1], [3, 2.2, -1]])    y,max_r = knnClassify(Target_feature=Target_feature,dataSet=dataSet,p=2,k=3)    print(y,max_r)
    fig = plt.figure()    plt.subplot(111)
    x_x1 = dataSet[dataSet[:, -1] == -1][:, :-1][:, 0].tolist()    x_x2 = dataSet[dataSet[:, -1] == -1][:, :-1][:, 1].tolist()    o_x1 = dataSet[dataSet[:, -1] == 1][:, :-1][:, 0].tolist()    o_x2 = dataSet[dataSet[:, -1] == 1][:, :-1][:, 1].tolist()
    plt.scatter(x_x1, x_x2, marker='x', s=20, c='red')    plt.scatter(o_x1, o_x2, marker='o', s=20, c='blue')
    # Draw Circle    a = np.arange(Target_feature[0] - max_r, Target_feature[0] + max_r, 0.0001)    b1 = np.sqrt(np.square(max_r) - np.square(a - Target_feature[0])) + Target_feature[1]    b2 = Target_feature[1] - np.sqrt(np.square(max_r) - np.square(a - Target_feature[0]))    plt.plot(a, b1, color='g', linestyle=':')    plt.plot(a, b2, color='g', linestyle=':')    plt.scatter(Target_feature[0], Target_feature[1], c='g', marker='o',s=200)
    plt.xlabel('X(1)')    plt.ylabel('X(2)')    plt.show()