Python機器學(xué)習(xí)k-近鄰算法(K Nearest Neighbor)實例詳解

# -*- coding:utf-8 -*-
#! python2
import numpy as np
import operator
# 訓(xùn)練集
data_set = np.array([[1., 1.1],
           [1.0, 1.0],
           [0., 0.],
           [0, 0.1]])
labels = ['A', 'A', 'B', 'B']
def classify_knn(in_vector, training_data, training_label, k):
  """
  :param in_vector: 待分類向量
  :param training_data: 訓(xùn)練集向量
  :param training_label: 訓(xùn)練集標簽
  :param k: 選擇最近鄰居的數(shù)目
  :return: 分類器對 in_vector 分類的類別
  """
  data_size = training_data.shape[0] # .shape[0] 返回二維數(shù)組的行數(shù)
  diff_mat = np.tile(in_vector, (data_size, 1)) - data_set # np.tile(array, (3, 2)) 對 array 進行 3×2 擴展為二維數(shù)組
  sq_diff_mat = diff_mat ** 2
  sq_distances = sq_diff_mat.sum(axis=1) # .sum(axis=1) 矩陣以列求和
  # distances = sq_distances ** 0.5 # 主要是通過比較求最近點,所以沒有必要求平方根
  distances_sorted_index = sq_distances.argsort() # .argsort() 對array進行排序 返回排序后對應(yīng)的索引
  class_count_dict = {} # 用于統(tǒng)計類別的個數(shù)
  for i in range(k):
    label = training_label[distances_sorted_index[i]]
    try:
      class_count_dict[label] += 1
    except KeyError:
      class_count_dict[label] = 1
  class_count_dict = sorted(class_count_dict.iteritems(), key=operator.itemgetter(1), reverse=True) # 根據(jù)字典的value值對字典進行逆序排序
  return class_count_dict[0][0]
if __name__ == '__main__':
  vector = [0, 0] # 待分類數(shù)據(jù)集
  print classify_knn(in_vector=vector, training_data=data_set, training_label=labels, k=3)