def sk():
datafile = './sutil/datasets/ex2data1.txt'
d = Dataset.fromDataFile(datafile, ',')
ms = LogisticRegression()
m = SklearnModel('Sklearn Logistic', ms)
m.trainModel(d)
m.score(d.X, d.y)
def test_normalize_features(self):
"""
Test for normalized distance function
"""
datafile = './sutil/datasets/ex1data1.txt'
d = Dataset.fromDataFile(datafile, ',')
print(d.shape)
def simple():
datafile = './sutil/datasets/ex2data1.txt'
d = Dataset.fromDataFile(datafile, ',')
theta = np.zeros((d.n + 1, 1))
lr = RegularizedLogisticRegression(theta, 0.03, 0)
lr.trainModel(d)
lr.score(d.X, d.y)
def test_save(self):
print("=" * 20)
print("Testing save")
datafile = './sutil/datasets/ex1data1.txt'
d = Dataset.fromDataFile(datafile, ',')
d.plotDataRegression('example')
d.save()
d.save('test')
print(d.shape)
def test_biased_x(self):
print("=" * 20)
print("Testing biased x")
datafile = './sutil/datasets/ex1data1.txt'
d = Dataset.fromDataFile(datafile, ',')
print(d.shape)
print(d.X[0])
print(d.normalizeFeatures())
print(d.getBiasedX())
def test_load_data(self):
"""
Test for the dataset from file function
"""
datafile = './sutil/datasets/ex1data1.txt'
d = Dataset.fromDataFile(datafile, ',')
d.plotDataRegression('example', False)
print(d.shape)
for i in range(len(d.X)):
print(str(d.X[i]) + ' -->' + str(d.y[i]))
datafile = './sutil/datasets/ex1data2.txt'
d2 = Dataset.fromDataFile(datafile, ',')
print(d2.shape)
datafile = './sutil/datasets/ex2data1.txt'
d3 = Dataset.fromDataFile(datafile, ',')
print(d3.shape)
d3.plotData('example3')
datafile = './sutil/datasets/ex2data2.txt'
d4 = Dataset.fromDataFile(datafile, ',')
print(d4.shape)
d4.plotData('example4')
def test_split(self):
print("=" * 20)
print("Testing split")
datafile = './sutil/datasets/ex1data1.txt'
d = Dataset.fromDataFile(datafile, ',')
d.plotDataRegression('example', True)
print(d.shape)
train, validation, test = d.split(0.8, 0.2)
print(train.shape, validation.shape, test.shape)
print(train.m, validation.m, test.m)
print(train.m / d.m, validation.m / d.m, test.m / d.m)
print(d.shape)
train1, test1 = d.split(0.8, 0)
print(train1.shape, test1.shape)
print(train1.m, test1.m)
print(train1.m / d.m, test1.m / d.m)
print(d.shape)
# -*- coding: utf-8 -*- import numpy as np from sutil.base.Dataset import Dataset from sutil.models.RegularizedLogisticRegression import RegularizedLogisticRegression from sutil.metrics.ModelROC import ModelROC datafile = './sutil/datasets/ex2data1.txt' d = Dataset.fromDataFile(datafile, ',') theta = np.zeros((d.n + 1, 1)) alpha = 0.03 l = 0 lr = RegularizedLogisticRegression(theta, alpha, l) lr.trainModel(d) m = ModelROC(lr, d.getBiasedX(), d.y, legend='Example of Model ROC usage') m.plot() m.zoom((0, 0.4), (0.5, 1.0))
def fromDataFile(cls, datafile, delimeter, alpha=0.1, l=0.1):
data = Dataset.fromDataFile(datafile, delimeter)
theta = np.random.random(data.X.shape[1] + 1)
return cls(data, theta, alpha, l)