IA 20211108

#generic_chi_j.py
#chi-s = (o-e)^2/e
import random


def chisquare(
    pListOfActualObservedValues:list,
    pListOfExpectedValues:list
)->float or bool:
    bCaution = len(pListOfActualObservedValues) \
                == len(pListOfExpectedValues)

    if (bCaution):
        fDistance=0
        iHowManySamples:int = len(pListOfActualObservedValues)
        listValidAddresses:list = \
            range(iHowManySamples) #0..iHowManySamples-1
        for iAddress in listValidAddresses:
            #compute
            observedValue = \
                pListOfActualObservedValues[iAddress]
            expectedValue = \
                pListOfExpectedValues[iAddress]
            fDiff = observedValue - expectedValue
            fParcelDist = fDiff**2
            fParcelChi = fParcelDist/expectedValue
            fDistance+=fParcelChi
        #for

        return fDistance
    else:
        return False
#def chisquare

#J(A,B) = |A∩B| / |A∪B|
#J(A,B) = card(intersect(A,B)) / card(union(A, B))
def jaccardProximity(
    pListOfActualObservedValues:list,
    pListOfExpectedValues:list
):
    #no repetitions
    setObserved = set(pListOfActualObservedValues)
    #type(setObserved) #set
    #no repetitions
    setExpected = set(pListOfExpectedValues)
    #type(setExpected) #set
    setIntersection = setObserved.intersection(setExpected)
    setUnion = setObserved.union(setExpected)
    cardinalityIntersect = len(setIntersection)
    cardinalityUnion = len(setUnion)
    fJProx = cardinalityIntersect / cardinalityUnion
    return fJProx
#def jaccardProximity

def randomListFloats(pLen:int):
    listRet = []
    for idx in range(pLen):
        f = random.Random()
        listRet.append(f)
    #for
    #[0.1 , 0.2323, ....] * 100 => [10.0, 23.23, ...[
    #broadcasting
    listRet*=100
    #listRet*=random.randint(10, 1000)
    #listRet = listRet*random.randint(10, 1000)
    return listRet
#def randomListFloats

col1 = randomListFloats(100)
print (col1)
col2 = randomListFloats(100)
print (col2)