Gaussian Distribution in more "Pythonic way"

Question

I am new to python coding . I have written a python code to calculate Gaussian distribution and predict the label of the set of values. This was my class assignment for which I got good marks. Now I want to know if my code is correct in more python sense. Can I do any more improvement to the code and make it precise and more "Pythonic".

import math
import operator 
# Class to get the mean  and variance of all data point. Input paramaters 
#are the Labels (M or W) and parameter to calculate (height, weight, age). 
#samples are the number of data points
def getMean(trainingSet,parameter1,parameter2):
    mean =0 
    samples = 0 
    variance = 0
    for x in range(len(trainingSet)):
        if trainingSet[x][3]==parameter2:
            mean+= trainingSet[x][parameter1]
            samples = samples+1
    finalMean = mean/samples
    #print(finalMean)
    for x in range(len(trainingSet)):
        if trainingSet[x][3]==parameter2:
            variance+= (trainingSet[x][parameter1]-finalMean)**2
    finalVariance = variance/samples
    gausVal = []
    for x in range(len(trainingSet)):
        tempval = 
calculateGuassian(finalMean,finalVariance,trainingSet[x][parameter1])
        gausVal.append(tempval)
    return gausVal

#Class to calculate the gussaian distriubion points 

def calculateGuassian(meanVal, varianceVal, feature1):
    DenoVariance = 2*varianceVal
    func1 = 1/(math.sqrt(2*3.14*varianceVal))
    func2 = (-(feature1-meanVal)**2)/DenoVariance
    func3 = math.exp(func2)
    distro = func1*func3
    return distro

def finalProduct(multiplyer):
    result = 1
    for x in multiplyer: 
        result = result*x
    return result   

def arrayMultiply(arr1, arr2) :
    resultArray = []
    for x in range(len(arr1)):
        arrMul = arr1[x]*arr2[x]
        resultArray.append(arrMul)
    return resultArray  


# Main classes where every feature is calculated multiplied and the result 
#is shown

def main() :
    MenArr = []
    WomenList = []
    heightM = getMean(trainSet,0,'M')
    finalHM = finalProduct(heightM)
    MenArr.append(finalHM)
    heightW = getMean(trainSet,0, 'W')
    finalHW = finalProduct(heightW)
    WomenList.append(finalHW)
    weightM = getMean(trainSet,1,'M')
    finalWM = finalProduct(weightM)
    MenArr.append(finalWM)
    weightW = getMean(trainSet,1,'W')
    finalWW = finalProduct(weightW)
    WomenList.append(finalWW)
    ageM = getMean(trainSet,2,'M')
    finalAM = finalProduct(ageM)
    MenArr.append(finalAM)
    ageW = getMean(trainSet,2,'W')
    finalAW = finalProduct(ageW)
    WomenList.append(finalAW)
    BestResultMTemp = arrayMultiply(MenArr,testData)
    BestResultWTemp = arrayMultiply(WomenList,testData)
    BestResultM = finalProduct(BestResultMTemp)*0.50
    BestResultW = finalProduct(BestResultWTemp)*0.50
    print (BestResultM)
    print(BestResultW)
    if BestResultM<BestResultW :
        print("The Class Label Is W")
    if BestResultM>BestResultW :
            print("The Class Label Is M")


trainSet = [[170, 57, 32, 'W'],
[192, 95, 28, 'M'],
[150, 45, 30, 'W'],
[170, 65, 29, 'M'],
[175, 78, 35, 'M'],
[185, 90, 32, 'M'],
[170, 65, 28, 'W'],
[155, 48, 31, 'W'],
[160, 55, 30, 'W'],
[182, 80, 30, 'M'],
[175, 69, 28, 'W'],
[180, 80, 27, 'M'],
[160, 50, 31, 'W'],
[175, 72, 30, 'M']]     
testData = (175, 70, 35)        
main()

Any kind of suggestion is most welcome. Thank You in advance.

Answer 1

Your question's title does not reflect what you are actually asking. You might be better fitting to http://codereview.stackexchange.com/ with this...

But, on first glance:

• Read and adhere to PEP8, which defines good Python coding practices: https://www.python.org/dev/peps/pep-0008/

• Maybe use things like autopep8 (https://pypi.org/project/autopep8/) or flake8 (http://flake8.pycqa.org/)

• Look up how if __name__ == "__main__": is used: What does if __name__ == "__main__": do?

• Look into NumPy (http://www.numpy.org/) specifically for things like your arrayMultiply() function.

• You are calling this a "class" in the header comment. However, there's no class defined anywhere here. So, look into what a Python class actually is: https://docs.python.org/3/tutorial/classes.html

Answer 2

As NichtJens mentioned, try to follow the PEP8 guide, so use lower case variable names with underscores to separate words, and add more spaces between characters.

Also try to use consistent and meaningful variable names. For example, why do you have MenArr and WomenList? You have a variable mean but its value is a sum. You have temporary variables with names func1, func2, etc..

To make your for loops more Pythonic loop through the items in a list rather than create an index and then look up the items.

So:

gauss_value = []
for x in range(len(lst)):
    value = calculate_guassian(mean, variance, lst[x][parameter1])
    gauss_value.append(value)
return gauss_value

You can do:

gauss_value = []
for item in lst:
    value = calculate_guassian(mean, variance, item[parameter1])
    gauss_value.append(value)
return gauss_value

But even better you can use a list comprehension:

gauss_value = [calculate_guassian(mean, variance, item[parameter1]) for item in lst]

You can use this to simplify a lot of your code, e.g. arrayMultiply could be:

def list_multiply(list_1, list_2) :
    return [a * b for a, b in zip(list_1, list_2)]

My version of getMean would filter the data first. I'm not sure if it's correct to use the unfiltered data for the calculateGuassian part:

def get_mean(values, index, label):
    filtered_values = [value[index] for value in values if value[3] == label]
    n = len(filtered_values)

    mean = sum(filtered_values) / n
    summed_squared_difference = sum((val - mean) ** 2 for val in filtered_values)
    variance = summed_squared_difference / n

    return [calculateGuassian(mean, variance, item[index]) for item in values]

You can also greatly reduce the amount of code required to get up the initial lists:

men_values = [product(get_mean(trainSet, i, 'M')) for i in range(3)]
women_values = [product(get_mean(trainSet, i, 'W')) for i in range(3)]

You could reduce the repetition further by having a function that take 'M' or 'W' as a parameter and returns the relevant list.