1. KNN iris案例

from sklearn.neighbors import KNeighborsClassifier
from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler

"""
    加载数据
"""
iris = datasets.load_iris()
X_data, y_data = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.25)

"""
    特征归一化
"""
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

"""
    训练
"""
clf = KNeighborsClassifier(n_neighbors=5)
clf.fit(X_train, y_train)

"""
    评估
"""
print("train score: ", clf.score(X_train, y_train))
print("test score: ", clf.score(X_test, y_test))

"""
    网格搜索
"""
param_grid = [
    {
        'weights': ['uniform', 'distance'],
        'n_neighbors': [i for i in range(3, 11)]
    }
]

clf = KNeighborsClassifier()
grid_search = GridSearchCV(clf, param_grid)
grid_search.fit(X_train, y_train)
print("Best set score: {:.2f}".format(grid_search.best_score_))
print("Best parameters: {}".format(grid_search.best_params_))
print("Test set score: {:.2f}".format(grid_search.score(X_test, y_test)))

2. KNeighborsClassifier参数说明

参考：

n_neighbors: int, optional (default = 5)

KNN的K

weights : str or callable, optional (default = “uniform”)

uniform：

p : integer, optional (default = 2)

p是metric的附属参数。在metric="manhattan"和metric="minkowski"时，指定对应的p参数

metric : string or DistanceMetric object (default=’minkowski’)

指定距离的计算方式

欧式距离 “euclidean”

曼哈顿距离 “manhattan”

切比雪夫距离 “chebyshev”

闵可夫斯基距离 “minkowski”

带权重闵可夫斯基距离 “wminkowski”

标准化欧式距离 “seuclidean”

马氏距离 “mahalanobis”