# coding: utf-8
# linear_regression/test_lwr.py
import regression
import matplotlib.pyplot as plt
import matplotlib.ticker as mtick
import numpy as np
if __name__ == "__main__":
srcX, y = regression.loadDataSet('data/lwr.txt')
m, n = srcX.shape
srcX = np.concatenate((srcX[:, 0], np.power(srcX[:, 0], 2)), axis=1)
# 特征缩放
X = regression.standardize(srcX.copy())
X = np.concatenate((np.ones((m, 1)), X), axis=1)
rate = 0.1
maxLoop = 1000
epsilon = 0.01
predicateX = regression.standardize(np.matrix([[8, 64]]))
predicateX = np.concatenate((np.ones((1, 1)), predicateX), axis=1)
result, t = regression.lwr(rate, maxLoop, epsilon, X, y, predicateX, 1)
theta, errors, thetas = result
result2, t = regression.lwr(rate, maxLoop, epsilon, X, y, predicateX, 0.1)
theta2, errors2, thetas2 = result2
# 打印特征点
# coding: utf-8
# linear_regression/test_temperature_polynomial.py
import regression
import matplotlib.pyplot as plt
import matplotlib.ticker as mtick
import numpy as np
if __name__ == "__main__":
srcX, y = regression.loadDataSet('temperature.txt')
m, n = srcX.shape
srcX = np.concatenate((srcX[:, 0], np.power(srcX[:, 0], 2)), axis=1)
# 特征缩放
X = regression.standardize(srcX.copy())
X = np.concatenate((np.ones((m, 1)), X), axis=1)
rate = 0.1
maxLoop = 1000
epsilon = 0.01
result, timeConsumed = regression.bgd(rate, maxLoop, epsilon, X, y)
theta, errors, thetas = result
# 打印特征点
fittingFig = plt.figure()
title = 'polynomial with bgd: rate=%.2f, maxLoop=%d, epsilon=%.3f \n time: %ds' % (
rate, maxLoop, epsilon, timeConsumed)
ax = fittingFig.add_subplot(111, title=title)
trainingSet = ax.scatter(srcX[:, 1].flatten().A[0], y[:, 0].flatten().A[0])
# linear_regression/test_temperature_normal.py
import regression as re
from matplotlib import cm
from mpl_toolkits.mplot3d import axes3d
import matplotlib.pyplot as plt
import matplotlib.ticker as mtick
import numpy as np
if __name__ == '__main__':
srcX, y = re.loadDataSet('data/temperature.txt');
m, n = srcX.shape
srcX = np.concatenate((srcX[:, 0], np.power(srcX[:, 0], 2)), axis=1)
# 特征缩放
X = re.standardize(srcX.copy())
X = np.concatenate((np.ones((m, 1)), X), axis=1)
rate = 0.1
maxLoop = 1000
epsilon = 0.01
result, timeConsumed = re.bgd(rate, maxLoop, epsilon, X, y)
theta, errors, thetas = result
# 打印特征点
fittingFig = plt.figure()
title = 'polynomial with bgd: rate=%.2f, maxLoop=%d, epsilon=%.3f \n time: %ds' % (rate, maxLoop, epsilon, timeConsumed)
ax = fittingFig.add_subplot(111, title=title)
trainingSet = ax.scatter(srcX[:, 0].flatten().A[0], y[:, 0].flatten().A[0])