from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
testName = "LR03-testLRCNN01-GA"
lr = NeuralNetworkCalculator(3, 10)
lr.load('nn_data.csv')
lr.runCnt = 100
lr.calcByGA(500, [-10.0, 10.0], 0.05)
print lr.betas
print lr.mse
lr.log(testName)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.fitnessVal)
#plt.show()
plt.savefig(testName)
print lr.fitnessVal[lr.runCnt-1]
from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
#lr = NeuralNetworkCalculator(7)
#lr.load('auto_mpg-norm.csv')
lr = NeuralNetworkCalculator(1)
lr.load('mlp_reg.csv')
#lr.calc()
lr.runCnt = 100
lr.calcByGA(1000, [-10.0, 10.0])
#lr.calcByPSO(1000, [-5.0, 5.0])
print lr.betas
print lr.mle
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.fitnessVal)
plt.show()
print lr.fitnessVal[lr.runCnt - 1]
from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
testName = "LR05-testLRCNN01-GA"
lr = NeuralNetworkCalculator(3, 10)
lr.loadTrainData('nn_data-train.csv')
lr.loadTestData('nn_data-test.csv')
lr.runCnt = 200
lr.calcByGA(1000, [-10.0, 10.0], 0.1)
lr.calcTestMSE(lr.betas)
print lr.betas
print "TRAIN SIZE: " + str(lr.trainDataSize) + " MSE: " + str(lr.trainMSE)
print "TEST SIZE: " + str(lr.testDataSize) + " MSE: " + str(lr.testMSE)
lr.log(testName)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.trainFitnessVal)
ax.set_xlabel("Iteration")
ax.set_ylabel("Fitness")
title_str = "Train M.S.E = " + str(lr.trainMSE) + " "
title_str += "Test M.S.E = " + str(lr.testMSE) + "\n"
title_str += " beta 0 = " + str(lr.betas[0])
title_str += ", beta 1 = " + str(lr.betas[1])
ax.set_title(title_str)
from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
testName = "LR02-testLRC07-GA"
lr = NeuralNetworkCalculator(6, 10)
lr.load('yacht-norm.csv')
lr.runCnt = 100
lr.calcByGA(1000, [-10.0, 10.0])
print lr.betas
print lr.mse
lr.log(testName)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.fitnessVal)
#plt.show()
plt.savefig(testName)
print lr.fitnessVal[lr.runCnt-1]
from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
testName = "LR02-testLRCNN01-PSO"
lr = NeuralNetworkCalculator(3)
lr.load('nn_data.csv')
lr.runCnt = 100
lr.calcByPSO(500, [-10.0, 10.0])
print lr.betas
print lr.mle
lr.log(testName)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.fitnessVal)
#plt.show()
plt.savefig(testName)
print lr.fitnessVal[lr.runCnt-1]
from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
testName = "LR02-testLRC06-GA"
lr = NeuralNetworkCalculator(7, 10)
lr.load('auto_mpg-norm.csv')
lr.runCnt = 500
lr.calcByGA(2000, [-10.0, 10.0])
print lr.betas
print lr.mse
lr.log(testName)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.fitnessVal)
#plt.show()
plt.savefig(testName)
print lr.fitnessVal[lr.runCnt - 1]
from NeuralNetworkCalculator import *
import matplotlib.pyplot as plt
if __name__ == '__main__':
testName = "LR02-testLRC06-PSO"
lr = NeuralNetworkCalculator(7)
lr.load('auto_mpg-norm.csv')
lr.runCnt = 100
lr.calcByPSO(500, [-10.0, 10.0])
print lr.betas
print lr.mle
lr.log(testName)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(np.arange(lr.runCnt), lr.fitnessVal)
#plt.show()
plt.savefig(testName)
print lr.fitnessVal[lr.runCnt-1]