AWD

import sys
#import numpy
#from pylab import *


def average(A):
    size = len(A)
    arraySum = 0
    for i in range(0, size):
        arraySum += A[i]
    return arraySum / size


def standard_deviation(A):
    avg = average(A)
    size = len(A)
    arraySum = 0
    for i in range(0, size):
        arraySum += ((A[i] - avg) * (A[i] - avg))
    return sqrt(arraySum / size)


def covariance(X, Y):
    avgX = average(X)
    avgY = average(Y)
    size = len(X)
    XY = []
    for i in range(0, size):
        XY.append(X[i] * Y[i])
    avgXY = average(XY)
    return avgXY - (avgX * avgY)


def rPearson_compound(X, Y):
    cov = covariance(X, Y)
    stdX = standard_deviation(X)
    stdY = standard_deviation(Y)
    return cov / (stdX * stdY)


def linear_regression(X, Y):
    cov = covariance(X, Y)
    stdX = standard_deviation(X)
    avgX = average(X)
    avgY = average(Y)
    a = cov / (stdX * stdX)
    b = avgY - a * avgX
    return {'a': a, 'b': b}


xArray = []
yArray = []
xyArray = []

with open(sys.argv[1], "r") as ins:
    for line in ins:
        splittedLine = line.split(" ")
        xArray.append(float(splittedLine[0]))
        yArray.append(float(splittedLine[1]))
        xyArray.append(float(splittedLine[0]) * float(splittedLine[1]))

xAvg = numpy.average(xArray)
yAvg = numpy.average(yArray)
xyAvg = numpy.average(xyArray)

xStd = numpy.std(xArray)
yStd = numpy.std(yArray)

# cov = numpy.cov([xArray, yArray], ddof = 0)

cov = xyAvg - (xAvg * yAvg);

cor = cov / (xStd * yStd)

regressionA = round(cov / (xStd * xStd), 3)
regressionB = round(yAvg - regressionA * xAvg, 3)

sm = min(xArray) - 1
bg = max(xArray) + 1

smY = regressionA * sm + regressionB
bgY = regressionA * bg + regressionB

lineX = [sm, bg]
lineY = [smY, bgY]

print(xArray);
print(yArray);
print("Średnia X: {}".format(xAvg))
print("Średnia Y: {}".format(yAvg))
print("Średnia X: {}".format(average(xArray)))
print("Średnia Y: {}".format(average(yArray)))
print("Odchylenie standardowe X: {}".format(xStd))
print("Odchylenie standardowe Y: {}".format(yStd))
print("Odchylenie standardowe X: {}".format(standard_deviation(xArray)))
print("Odchylenie standardowe Y: {}".format(standard_deviation(yArray)))
print("Kowariancja: {}".format(cov))
print("Kowariancja: {}".format(covariance(xArray, yArray)))
print("Współczynnik korelacji r-Pearsona: {}".format(cor))
print("Współczynnik korelacji r-Pearsona: {}".format(rPearson_compound(xArray, yArray)))
print("Prosta regresji : y = {} * x + {}".format(regressionA, regressionB))
reg = linear_regression(xArray, yArray)
print("Prosta regresji : y = {} * x + {}".format(reg['a'], reg['b']))

plot(xArray, yArray, "ro");
plot(lineX, lineY)
xlabel("Oś x")
ylabel("Oś y")
title("AWD")
savefig("awd.png")
print("Wykres zapisano do pliku awd.png")