def main():
fileW = createAnOutputFile()
model = svm.SVR()
numOfPop = 50 # should be 200 population, lower is faster but less accurate
numOfFea = 385 # should be 385 descriptors
unfit = 1000
# Final model requirements
R2req_train = .6
R2req_validate = .5
R2req_test = .5
TrainX, TrainY, ValidateX, ValidateY, TestX, TestY = FromDataFileSVM_GA.getAllOfTheData()
TrainX, ValidateX, TestX = FromDataFileSVM_GA.rescaleTheData(TrainX, ValidateX, TestX)
#numOfFea = TrainX.shape[1] # should be 396 descriptors
unfit = 1000 # when to stop when the model isn't doing well
fittingStatus = unfit
while (fittingStatus == unfit):
population = createInitialPopulation(numOfPop, numOfFea)
fittingStatus, fitness = FromFinessFileSVM_GA.validate_model(model, fileW, population,
TrainX, TrainY, ValidateX, ValidateY, TestX, TestY)
elite1, elite2, elite1Index, elite2Index = findElites(fitness, population)
sumOfFitnesses = AddFitnessForFindingParents(fitness)
print "Starting the Loop - time is = ", time.strftime("%H:%M:%S", time.localtime())
IterateNtimes(model, fileW, fitness, sumOfFitnesses, population,
elite1, elite2, elite1Index, elite2Index,
TrainX, TrainY, ValidateX, ValidateY, TestX, TestY)
return
def IterateNtimes(model, fileW, fitness, sumOfFitnesses, population,
elite1, elite2, elite1Index, elite2Index,
TrainX, TrainY, ValidateX, ValidateY, TestX, TestY):
unfit = 1000
numOfGenerations = 2000 # should be 2000
numOfPop = population.shape[0]
numOfFea = population.shape[1]
Times = 0
oldFitness = fitness.min()
for i in range(1, numOfGenerations):
oldFitness, Times = checkterTerminationStatus(Times, oldFitness, fitness.min())
print "This is generation ", i, " --- Minimum of fitness is: -> ", fitness.min()
if (fitness.min() < 0.005):
print "***********************************"
print "Good: Fitness is low enough to quit"
print "***********************************"
exit(0)
fittingStatus = unfit
while (fittingStatus == unfit):
population = findNewPopulation(elite1, elite2, sumOfFitnesses, population)
fittingStatus, fitness = FromFinessFileSVM_GA.validate_model(model, fileW,
population, TrainX, TrainY, ValidateX,
ValidateY, TestX, TestY)
elite1, elite2, elite1Index, elite2Index = findElites(fitness, population)
#adding all the fitnesses and storing them in one dimensional array for
#choosing the children for the next round
sumFitnesses = AddFitnessForFindingParents(fitness)
return
