import numpy as np
from matplotlib.lines import Line2D
from MachineLearn.Classes import Experiment
import matplotlib.pyplot as plt
oExp11 = Experiment.load("Objects/EXP01_1_LP_20.gzip".format())
oExp12 = Experiment.load("Objects/EXP01_2_LP_20.gzip".format())
oExp13 = Experiment.load("Objects/EXP01_3_LP_20.gzip".format())
oExp14 = Experiment.load("Objects/EXP01_4_LP_20.gzip".format())
oExp15 = Experiment.load("Objects/EXP01_5_LP_20.gzip".format())
COLORS = ['GREEN', 'RED', 'BLUE']
MARKER = ['o', '^', "*"]
base1 = np.loadtxt("Datasets/XOR.txt", delimiter=",")
print(oExp11)
print()
print(oExp12)
print()
print(oExp13)
print()
print(oExp14)
print()
print(oExp15)
print()
def getBestTrain(exp, name):
"""Etapa 1: Matriz confusao e grafico para melhor treinamento."""
oDataSet = exp.experimentResults[0]
oDataSet.add_sample_of_attribute(
np.array(list(np.float32(y)) + [classes[x]]))
oDataSet.attributes = oDataSet.attributes.astype(float)
oDataSet.normalize_data_set()
for j in range(20):
print(j)
oData = Data(len(oDataSet.labelsNames), 31, samples=50)
oData.random_training_test_by_percent(
np.unique(classes, return_counts=True)[1], 0.8)
perc = Layered_perceptron_Logistic(learning_rate,
len(oDataSet.labelsNames))
perc.train(oDataSet.attributes[oData.Training_indexes],
oDataSet.labels[oData.Training_indexes], epochs)
oData.model = perc
oData.confusion_matrix = np.zeros(
(len(oDataSet.labelsNames), len(oDataSet.labelsNames)))
for i in oData.Testing_indexes:
data = np.matrix(np.hstack(([-1], oDataSet.attributes[i]))).T
predicted = perc.predict(data)
oData.confusion_matrix[int(oDataSet.labels[i]), predicted] += 1
print(oData)
oDataSet.append(oData)
oExp.add_data_set(
oDataSet,
description=" Experimento Dermatologia LP 20 realizaçoes.".format())
oExp.save("Objects/EXP01_4_LP_20.gzip".format())
oExp = Experiment.load("Objects/EXP01_4_LP_20.gzip".format())
print(oExp)
print(oExp.experimentResults[0].sum_confusion_matrix)
import numpy as np
from matplotlib.lines import Line2D
from MachineLearn.Classes import Experiment
import matplotlib.pyplot as plt
oExp11 = Experiment.load("Objects/EXP02_1_LP_20.gzip".format())
oExp12 = Experiment.load("Objects/EXP02_2_LP_20.gzip".format())
oExp13 = Experiment.load("Objects/EXP02_3_LP_20.gzip".format())
COLORS = ['GREEN', 'RED', 'BLUE']
MARKER = ['o', '^', "*"]
base1 = np.loadtxt("Datasets/artifitial1.data", delimiter=",")
def imprimir_resultado(oexp, name, oData):
oDataSet = oexp.experimentResults[0]
print(
"EXPERIMENTO " + name + " MELHOR RESULTADO MSE: ",
oData.params['MSE'],
)
RMSE = []
MSE = []
for i in oDataSet.dataSet:
RMSE.append(i.params['RMSE'])
MSE.append(i.params['MSE'])
MSE = np.array(MSE)
RMSE = np.array(RMSE)
print("\tRMSE MEDIO ", np.mean(RMSE), "DESVIO", np.std(RMSE))
print("\tMSE MEDIO ", np.mean(MSE), "DESVIO", np.std(MSE))
oDataSet = DataSet()
base = np.loadtxt("Datasets/dataset2.txt", usecols=range(2), delimiter=",")
classes = np.loadtxt("Datasets/dataset2.txt", usecols=-1, delimiter=",")
for x, y in enumerate(base):
oDataSet.add_sample_of_attribute(
np.array(list(np.float32(y)) + [classes[x]]))
oDataSet.attributes = oDataSet.attributes.astype(float)
# oDataSet.normalize_data_set()
for j in range(20):
print(j)
oData = Data(2, 31, samples=50)
oData.random_training_test_by_percent([600, 600], 0.8)
perc = Perceptron(learning_rate)
perc.train(oDataSet.attributes[oData.Training_indexes],
oDataSet.labels[oData.Training_indexes], epochs)
oData.model = perc
oData.confusion_matrix = np.zeros((2, 2))
for i in oData.Testing_indexes:
data = np.matrix(np.hstack(([-1], oDataSet.attributes[i]))).T
oData.confusion_matrix[int(oDataSet.labels[i]),
perc.predict(data)] += 1
oDataSet.append(oData)
oExp.add_data_set(
oDataSet, description=" Experimento dataset2 PS 20 realizaçoes.".format())
oExp.save("Objects/EXP02_PS_20.gzip".format())
oExp = Experiment.load("Objects/EXP02_PS_20.gzip".format())
print(oExp)
print(oExp.experimentResults[0].sum_confusion_matrix)
import numpy as np
from matplotlib.lines import Line2D
from MachineLearn.Classes import Experiment
import matplotlib.pyplot as plt
oExp11 = Experiment.load("Objects/EXP01_DT1_20.gzip".format())
COLORS = ['GREEN', 'RED', 'BLUE']
base1 = np.loadtxt("Datasets/dt_1.txt", delimiter=" ")
# Etapa 1
def getBestTrain(exp):
"""Etapa 1: Matriz confusao e grafico para melhor treinamento."""
oDataSet = exp.experimentResults[0]
best = 1000000
oBestData = None
for oData in oDataSet.dataSet:
txAcc = oData.params['MSE']
if txAcc < best:
best = txAcc
oBestData = oData
return oBestData
oData11 = getBestTrain(oExp11)
oDataSet11 = oExp11.experimentResults[0]
print("EXPERIMENTO 1 MELHOR RESULTADO", oData11.params)
RMSE = []
MSE = []
import numpy as np
from matplotlib.lines import Line2D
from MachineLearn.Classes import Experiment
import matplotlib.pyplot as plt
oExp11 = Experiment.load("Objects/EXP01_1_PS_20.gzip".format())
oExp12 = Experiment.load("Objects/EXP01_2_PS_20.gzip".format())
oExp13 = Experiment.load("Objects/EXP01_3_PS_20.gzip".format())
oExp2 = Experiment.load("Objects/EXP02_PS_20.gzip".format())
COLORS = ['RED', 'BLUE']
print(oExp11)
print()
print(oExp12)
print()
print(oExp13)
print()
print(oExp2)
print()
# Etapa 1
def getBestTrain(exp, name):
"""Etapa 1: Matriz confusao e grafico para melhor treinamento."""
oDataSet = exp.experimentResults[0]
best = 0
oBestData = None
for oData in oDataSet.dataSet:
txAcc = oData.get_metrics()[1, -1]
if txAcc > best: