logistics regression:
(梯度下降法)
code:
import numpy as np
import random
def gradientDescent(x,y,theta,alpha,m,numIterations):
xTrans = x.transpose()
for i in range(0,numIterations):
hypothesis = np.dot(x,theta)
loss = hypothesis - y
cost = np.sum(loss ** 2)/(2 * m)
print("Iteration %d / Cost: %f" %(i,cost))
gradient = np.dot(xTrans,loss)/m
theta = theta - alpha*gradient
return theta
def genData(numPoints,bias,variance):
x = np.zeros(shape=(numPoints,2))
y = np.zeros(shape=numPoints)
for i in range(0,numPoints):
x[i][0] = 1
x[i][1] = i
y[i] = (i+bias)+random.uniform(0,1)*variance
return x,y
x,y = genData(100,25,10)
print("x:")
print(x)
print("y:")
print(y)
m,n = np.shape(x)
n_y = np.shape(y)
print("x shape:",str(m)," ",str(n))
print("y shape:",str(n_y))
numIterations = 100000
alpha = 0.0005
theta = np.ones(n)
theta = gradientDescent(x,y,theta,alpha,m,numIterations)
print(theta)
results:
...
Iteration 99990 / Cost: 3.645522
Iteration 99991 / Cost: 3.645522
Iteration 99992 / Cost: 3.645522
Iteration 99993 / Cost: 3.645522
Iteration 99994 / Cost: 3.645522
Iteration 99995 / Cost: 3.645522
Iteration 99996 / Cost: 3.645522
Iteration 99997 / Cost: 3.645522
Iteration 99998 / Cost: 3.645522
Iteration 99999 / Cost: 3.645522
[ 29.66956034 1.00586986]