2019-11-10

测试代码

```

import numpy as np

from math import sqrt

from collections import Counter

#from .metrics import accuracy_score

class kNNClassifier:

    def __init__(self,k):

        """初始化分类器"""

        assert k>=1,"k must be valid"

        self.k = k

        self._X_train = None

        self._y_train = None

    def fit(self,X_train,y_train):

        """根据训练数据集X_train和y_train训练knn分类器kNN分类器"""

        assert X_train.shape[0] == y_train.shape[0],\

        "the size of X_train must be equal to the size of y_train"

        assert self.k <= X_train.shape[0],\

        "the size of X_train must be at least k"

        self._X_train = X_train

        self._y_train = y_train

        return self

    def predict(self,X_predict):

        """给定待遇测数据集X_predcit,返回表示X_predict结果的向量"""

        assert self._X_train is not None and self._y_train is not None,\

        "must fit before predict!"

        assert X_predict.shape[1] == self._X_train.shape[1],\

        "the feature number of X_predict must be equal to X_train"

        y_predict = [self._predict(x) for x in X_predict]

        return np.array(y_predict)

    def _predict(self,x):

        distances = [sqrt(np.sum((x_train - x)**2)) for x_train in self._X_train]

        nearest = np.argsort(distances)

        topK_y = [self._y_train[i] for i in nearest]

        votes = Counter(topK_y)

        return votes.most_common(1)[0][0] 

    def __repr__(self):

        return "kNN(k=%d)" %self.k

```