# else
# point[c]=c1
# c1=avg(X where point[c]=c1)
# c2=avg(X where point[c]=c2)
#pyspark --packages com.databricks:spark-csv_2.10:1.3.0
import numpy as np
import random
from pyspark.sql import SQLContext
sqlContext = SQLContext(sc)
#first argument is data file name
#path=sys.argv[1]
dataFilePath="train.csv"
trainDataFrame = sqlContext.csvFile(source="com.databricks.spark.csv", header="true", path = dataFilePath)
trainDataFrame.registerTempTable("trainDataFrame")
def isCentroid(point, centroidArray):
i=1
for centroid in centroidArray:
if(("%.6f" %point[0]=="%.6f" %centroid[1][0]) & ("%.6f" %point[1]=="%.6f" %centroid[1][1])):
return "C"+str(i)
i=i+1
return "Not Centroid"
def getCentroid(point, centroidArray):
res=isCentroid(point, centroidArray)
if(res=="Not Centroid"):
myCentroid=centroidArray[0]
