python knn代码详解_具体knn算法概念参考knn代码python实现

具体knn算法概念参考knn代码python实现

上面是参考《机器学习实战》的代码，和knn的思想python

# _*_ encoding=utf8 _*_算法

import numpy as np

import tensorflow as tf

from tensorflow.examples.tutorials.mnist import input_data机器学习

# 导入手写体识别的数据

mnist = input_data.read_data_sets("../data", one_hot=True)post

# 训练集和测试集

X_train, Y_train = mnist.train.next_batch(5000) # 数据和labels

X_test, Y_test = mnist.test.next_batch(100)学习

# 定义输入

x_train = tf.placeholder(tf.float32, shape=(None,784))

x_test = tf.placeholder(tf.float32, shape=(784))测试

# L1距离也就是城市街区距离 |x1-x2|+|y1-y2|

distance = tf.reduce_sum(tf.abs(www.leyou1178.cn/   tf.add(x_train,tf.negative(x_test))),reduction_indices = 1)blog

# 返回最近的坐标，0纵轴 1横轴

pred = tf.arg_min(distance, 0)索引

accuracy = 0input

# 初始化

init = tf.global_variables_initializer()it

with tf.Session() as sess:

sess.run(init)

for i in range(len(X_test)):

# 获取当前样本的最近邻索引，当前样本和每个训练的样本找一个最近的l1距离，获得这个最小距离的下标

nn_index = sess.run(pred, feed_dict={x_train:X_train, x_test: X_test[i, :]})

# 由最邻近索引找到label,而后最邻近的label与真实标签比较 np.argmax找最大的下标

# 由l1距离找到的最小值对应的坐标，经过该最坐标找到对应行label的最大值的下标，这个下标对应的就是数字的大小

print("预测次数", i, "预测标签:", np.argmax(Y_train[nn_index]),"真实标签:", np.argmax(Y_test[i]))

# 计算准确率

if np.argmax(Y_www.xinghenyule.com train[nn_index]) www.089188.cn/==www.tianzunyule178.com  np.argmax(Y_test[i]):

accuracy += 1

print("Accuracy:", float(accuracy)/len(X_test))