面试手撕KNN算法

KNN模型没有训练过程，只需要用新数据与train数据计算距离即可。

直接上代码

# 参考
# https://zhuanlan.zhihu.com/p/59755939
# https://zhuanlan.zhihu.com/p/165130369

from collections import Counter
from sklearn import datasets
from sklearn.model_selection import train_test_split
import numpy as np

iris = datasets.load_iris()
X, y = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1234)


class KNN:
    def __init__(self, k = 3):
        self.k = k
        
    # 训练数据
    def fit(self, X, y):
        self.X_train = X
        self.y_train = y
        
    # 预测
    def predict(self, X_test):
        predicted_labels = [self._predict(x) for x in X_test]
        return np.array(predicted_labels)

    def _predict(self, x):
        # 计算1个新数据与所有训练数据的距离
        distances = [np.sqrt(np.sum((x_train - x) ** 2)) for x_train in self.X_train]
        
        # 获取 k 邻近样本和标签
        k_indices = np.argsort(distances)[:self.k]
        k_nearest_labels = [self.y_train[i] for i in k_indices]
        
        # 通过投票进行分类
        most_common = Counter(k_nearest_labels).most_common(1)
        return most_common[0][0]


if __name__ == "__main__":
    model = KNN(k = 3)
    model.fit(X_train, y_train)
    predictions = model.predict(X_test)
    acc = np.sum(predictions == y_test) / len(y_test)
    print(acc)