#!/usr/bin/python
import operator
import math

def euclideanDistance(data_set1, data_set2, length):
    distance = 0
    for x in range(length):
        distance += pow((data_set1[x] - data_set2[x]), 2)
    return math.sqrt(distance)

#k-nn regression (k-nearest neighbors algorithm) using the euclidean distance
def knnR(training_set, test_set, k, length, colTarget):

    distances = []
    avg = 0
    sum_avg = 0

    for x in range(len(training_set)):
        dist = euclideanDistance(test_set, training_set[x], length)
        distances.append((training_set[x], dist))
    distances.sort(key=operator.itemgetter(1))
    neighbors = []
    for x in range(k):
        neighbors.append(distances[x][0])

    #Average calculation
    for n in range(len(neighbors)):
        sum_avg += neighbors[n][colTarget]
    avg = sum_avg/len(neighbors)

    return avg