Created
August 8, 2018 08:28
-
-
Save ababycat/483fd68e08388537b5d95319d5f4b442 to your computer and use it in GitHub Desktop.
EM(Expectation-Maimization) algorithm for find multi Gaussian means.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
import matplotlib.pyplot as plt | |
SIGMA = 1 | |
def gauss(x, mu, sigma, eps=1e-8): | |
return 1./np.sqrt(np.pi*2*sigma**2+eps) * np.exp(-(x-mu)**2/(2*sigma**2+eps)) | |
def gauss_vec(x, mu, sigma, eps=1e-8): | |
""" x: Nx1 | |
mu: 1xm""" | |
x = x.reshape(-1, 1) | |
mu = mu.reshape(1, -1) | |
x = x.repeat(mu.shape[1], axis=1) | |
mu = mu.repeat(x.shape[0], axis=0) | |
return gauss(x, mu, sigma, eps) | |
def init_h_of_mu(k, begin, stop): | |
return (np.random.rand(k)*(stop-begin)+begin).reshape(1, -1) | |
def step1_calculate_Expectation(x, h_of_mu): | |
# posibility | |
p = gauss_vec(x, h, SIGMA) | |
p = p/np.sum(p, axis=1, keepdims=True) | |
return p | |
def step2_calculate_h_of_mu(x, expetation): | |
return np.sum(x * expetation, axis=0)/np.sum(expetation, axis=0) | |
def get_multi_gauss(mu=[1, 7, 15]): | |
samples = 200 | |
mu = np.array([mu]).reshape(1, -1) | |
mu = mu.repeat(samples, axis=0) | |
ds = np.random.randn(samples, mu.shape[1])*SIGMA | |
ds += np.random.randn(*ds.shape)*0.01 | |
return (ds + mu).reshape(-1) | |
t = get_multi_gauss([1, 7, 20]) | |
plt.plot(t, np.zeros_like(t), '.') | |
k = 3 | |
h = init_h_of_mu(k, -20, 20) | |
print('init state', h) | |
for i in range(8): | |
p = step1_calculate_Expectation(t.reshape(-1, 1), h) | |
h = step2_calculate_h_of_mu(t.reshape(-1, 1), p) | |
print(h) | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment