本文之編寫程序涉及到API介紹绵脯，程序的完整實(shí)現(xiàn)，具體算法原理請(qǐng)查看之前所寫的KNN算法介紹

一、基礎(chǔ)準(zhǔn)備

1嫌套、python 基礎(chǔ)

2、numpy 基礎(chǔ)

np.mean
求平均值

print(np.mean([1,2,3,4]))
# >> 2.5

3圾旨、scikit 基礎(chǔ)

fit
(X, y)
符合模型使用X作為訓(xùn)練數(shù)據(jù)和y值作為目標(biāo)
get_params
([deep])
得到的參數(shù)估計(jì)量踱讨。
.

kneighbors
([X, n_neighbors, return_distance])
發(fā)現(xiàn)的K-neighbors點(diǎn)。

kneighbors_graph
([X, n_neighbors, mode])
計(jì)算(加權(quán))圖k-Neighbors X點(diǎn)

predict
(X)
預(yù)測(cè)類標(biāo)簽所提供的數(shù)據(jù)

predict_proba
(X)
回歸測(cè)試數(shù)據(jù)的概率估計(jì)X砍的。

score
(X, y[, sample_weight])
返回意味著在給定的精度測(cè)試數(shù)據(jù)和標(biāo)簽痹筛。

set_params
(**params)
設(shè)置的參數(shù)估計(jì)量。
.

二廓鞠、完整程序

# -*- coding: utf-8 -*-
import numpy as np
from sklearn import neighbors, preprocessing
from sklearn.metrics import precision_recall_curve
from sklearn.metrics import classification_report
from sklearn.cross_validation import train_test_split

def file2Mat(testFileName, parammterNumber):
    fr = open(testFileName)
    lines = fr.readlines()
    lineNums = len(lines)
    resultMat = np.zeros((lineNums, parammterNumber))
    classLabelVector = []
    for i in range(lineNums):
        line = lines[i].strip()
        itemMat = line.split('\t')
        resultMat[i, :] = itemMat[0:parammterNumber]
        classLabelVector.append(itemMat[-1])
    fr.close()
    return resultMat, classLabelVector;


# 為了防止某個(gè)屬性對(duì)結(jié)果產(chǎn)生很大的影響帚稠，所以有了這個(gè)優(yōu)化，比如:10000,4.5,6.8 10000就對(duì)結(jié)果基本起了決定作用
def autoNorm(dataSet):
    minVals = dataSet.min(0)
    maxVals = dataSet.max(0)
    ranges = maxVals - minVals
    normMat = np.zeros(np.shape(dataSet))
    size = normMat.shape[0]
    normMat = dataSet - np.tile(minVals, (size, 1))
    normMat = normMat / np.tile(ranges, (size, 1))
    return normMat, minVals, ranges

if __name__=='__main__':
    trainigSetFileName = 'data\\datingTrainingSet.txt'
    testFileName = 'data\\datingTestSet.txt'

    # 讀取訓(xùn)練數(shù)據(jù)
    trianingMat, classLabel = file2Mat(trainigSetFileName, 3)
    # 對(duì)數(shù)據(jù)進(jìn)行歸一化的處理
    autoNormTrianingMat, minVals, ranges = autoNorm(trianingMat)
    # 讀取測(cè)試數(shù)據(jù)
    testMat, testLabel = file2Mat(testFileName, 3)
    autoNormTestMat = []
    for i in range(len(testLabel)):
        autoNormTestMat.append( (testMat[i] - minVals) / ranges)
    # testMat = preprocessing.normalize(testMat)
    print(autoNormTestMat)
    # ''''' 訓(xùn)練KNN分類器 '''
    clf = neighbors.KNeighborsClassifier(n_neighbors=5, algorithm='kd_tree')
    clf.fit(autoNormTrianingMat, classLabel)

    answer = clf.predict(autoNormTestMat)

    print(np.sum(answer != testLabel))

    # 計(jì)算分?jǐn)?shù)
    print(clf.score(autoNormTestMat, testLabel))
    print(np.mean(answer == testLabel))
    print(clf.predict([0.44832535,  0.39805139,  0.56233353]))
    print(clf.predict_proba([0.44832535,  0.39805139,  0.56233353]))
    # '''''準(zhǔn)確率與召回率'''
    # precision, recall, thresholds = precision_recall_curve(testLabel, clf.predict(testMat))