1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307
|
#! /usr/bin/python
#
# Copyrighted David Cournapeau
# Last Change: Thu Jul 12 04:00 PM 2007 J
"""This module implements various basic functions related to multivariate
gaussian, such as pdf estimation, confidence interval/ellipsoids, etc..."""
__docformat__ = 'restructuredtext'
import numpy as N
import numpy.linalg as lin
#from numpy.random import randn
from scipy.stats import chi2
import misc
# Error classes
class DenError(Exception):
"""Base class for exceptions in this module.
Attributes:
expression -- input expression in which the error occurred
message -- explanation of the error"""
def __init__(self, message):
self.message = message
Exception.__init__(self)
def __str__(self):
return self.message
# The following function do all the fancy stuff to check that parameters
# are Ok, and call the right implementation if args are OK.
def gauss_den(x, mu, va, log = False):
"""Compute multivariate Gaussian density at points x for
mean mu and variance va.
:Parameters:
x : ndarray
points where to estimate the pdf. each row of the array is one
point of d dimension
mu : ndarray
mean of the pdf. Should have same dimension d than points in x.
va : ndarray
variance of the pdf. If va has d elements, va is interpreted as the
diagonal elements of the actual covariance matrix. Otherwise,
should be a dxd matrix (and positive definite).
log : boolean
if True, returns the log-pdf instead of the pdf.
:Returns:
pdf : ndarray
Returns a rank 1 array of the pdf at points x.
Note
----
Vector are row vectors, except va which can be a matrix
(row vector variance for diagonal variance)."""
lmu = N.atleast_2d(mu)
lva = N.atleast_2d(va)
lx = N.atleast_2d(x)
#=======================#
# Checking parameters #
#=======================#
if len(N.shape(lmu)) != 2:
raise DenError("mu is not rank 2")
if len(N.shape(lva)) != 2:
raise DenError("va is not rank 2")
if len(N.shape(lx)) != 2:
raise DenError("x is not rank 2")
d = N.shape(lx)[1]
(dm0, dm1) = N.shape(lmu)
(dv0, dv1) = N.shape(lva)
# Check x and mu same dimension
if dm0 != 1:
msg = "mean must be a row vector!"
raise DenError(msg)
if dm1 != d:
msg = "x and mu not same dim"
raise DenError(msg)
# Check va and mu same size
if dv1 != d:
msg = "mu and va not same dim"
raise DenError(msg)
if dv0 != 1 and dv0 != d:
msg = "va not square"
raise DenError(msg)
#===============#
# Computation #
#===============#
if d == 1:
# scalar case
return _scalar_gauss_den(lx[:, 0], lmu[0, 0], lva[0, 0], log)
elif dv0 == 1:
# Diagonal matrix case
return _diag_gauss_den(lx, lmu, lva, log)
elif dv1 == dv0:
# full case
return _full_gauss_den(lx, lmu, lva, log)
else:
raise DenError("variance mode not recognized, this is a bug")
# Those 3 functions do almost all the actual computation
def _scalar_gauss_den(x, mu, va, log):
""" This function is the actual implementation
of gaussian pdf in scalar case. It assumes all args
are conformant, so it should not be used directly
Call gauss_den instead"""
d = mu.size
inva = 1/va
fac = (2*N.pi) ** (-d/2.0) * N.sqrt(inva)
inva *= -0.5
y = ((x-mu) ** 2) * inva
if not log:
y = fac * N.exp(y)
else:
y += N.log(fac)
return y
def _diag_gauss_den(x, mu, va, log):
""" This function is the actual implementation
of gaussian pdf in scalar case. It assumes all args
are conformant, so it should not be used directly
Call gauss_den instead"""
# Diagonal matrix case
d = mu.size
#n = x.shape[0]
if not log:
inva = 1/va[0]
fac = (2*N.pi) ** (-d/2.0) * N.prod(N.sqrt(inva))
inva *= -0.5
x = x - mu
x **= 2
y = fac * N.exp(N.dot(x, inva))
else:
# XXX optimize log case as non log case above
y = _scalar_gauss_den(x[:, 0], mu[0, 0], va[0, 0], log)
for i in range(1, d):
y += _scalar_gauss_den(x[:, i], mu[0, i], va[0, i], log)
return y
def _full_gauss_den(x, mu, va, log):
""" This function is the actual implementation
of gaussian pdf in full matrix case.
It assumes all args are conformant, so it should
not be used directly Call gauss_den instead
Does not check if va is definite positive (on inversible
for that matter), so the inverse computation and/or determinant
would throw an exception."""
d = mu.size
inva = lin.inv(va)
fac = 1 / N.sqrt( (2*N.pi) ** d * N.fabs(lin.det(va)))
# we are using a trick with sum to "emulate"
# the matrix multiplication inva * x without any explicit loop
#y = -0.5 * N.sum(N.dot((x-mu), inva) * (x-mu), 1)
y = -0.5 * N.dot(N.dot((x-mu), inva) * (x-mu),
N.ones((mu.size, 1), x.dtype))[:, 0]
if not log:
y = fac * N.exp(y)
else:
y = y + N.log(fac)
return y
# To get coordinatea of a confidence ellipse from multi-variate gaussian pdf
def gauss_ell(mu, va, dim = misc.DEF_VIS_DIM, npoints = misc.DEF_ELL_NP, \
level = misc.DEF_LEVEL):
"""Given a mean and covariance for multi-variate
gaussian, returns the coordinates of the confidense ellipsoid.
Compute npoints coordinates for the ellipse of confidence of given level
(all points will be inside the ellipsoides with a probability equal to
level).
:Parameters:
mu : ndarray
mean of the pdf
va : ndarray
variance of the pdf
dim : sequence
sequences of two integers which represent the dimensions where to
project the ellipsoid.
npoints: int
number of points to generate for the ellipse.
level : float
level of confidence (between 0 and 1).
:Returns:
Returns the coordinate x and y of the ellipse."""
if level >= 1 or level <= 0:
raise ValueError("level should be a scale strictly between 0 and 1.""")
mu = N.atleast_1d(mu)
va = N.atleast_1d(va)
d = N.shape(mu)[0]
c = N.array(dim)
if N.any(c < 0) or N.any(c >= d):
raise ValueError("dim elements should be >= 0 and < %d (dimension"\
" of the variance)" % d)
if N.size(mu) == N.size(va):
mode = 'diag'
else:
if N.ndim(va) == 2:
if N.shape(va)[0] == N.shape(va)[1]:
mode = 'full'
else:
raise DenError("variance not square")
else:
raise DenError("mean and variance are not dim conformant")
# When X is a sample from multivariante N(mu, sigma), (X-mu)Sigma^-1(X-mu)
# follows a Chi2(d) law. Here, we only take 2 dimension, so Chi2 with 2
# degree of freedom (See Wasserman. This is easy to see with characteristic
# functions)
chi22d = chi2(2)
mahal = N.sqrt(chi22d.ppf(level))
# Generates a circle of npoints
theta = N.linspace(0, 2 * N.pi, npoints)
circle = mahal * N.array([N.cos(theta), N.sin(theta)])
# Get the dimension which we are interested in:
mu = mu[c]
if mode == 'diag':
va = va[c]
elps = N.outer(mu, N.ones(npoints))
elps += N.dot(N.diag(N.sqrt(va)), circle)
elif mode == 'full':
va = va[c, :][:, c]
# Method: compute the cholesky decomp of each cov matrix, that is
# compute cova such as va = cova * cova'
# WARN: scipy is different than matlab here, as scipy computes a lower
# triangular cholesky decomp:
# - va = cova * cova' (scipy)
# - va = cova' * cova (matlab)
# So take care when comparing results with matlab !
cova = lin.cholesky(va)
elps = N.outer(mu, N.ones(npoints))
elps += N.dot(cova, circle)
else:
raise ValueError("var mode not recognized")
return elps[0, :], elps[1, :]
def logsumexp(x):
"""Compute log(sum(exp(x), 1)) while avoiding underflow.
:Parameters:
x : ndarray
data in log domain to sum"""
axis = 1
mc = N.max(x, axis)
return mc + N.log(N.sum(N.exp(x-mc[:, N.newaxis]), axis))
def multiple_gauss_den(data, mu, va, log = False):
"""Helper function to generate several Gaussian
pdf (different parameters) at the same points
:Parameters:
data : ndarray
points where to estimate the pdfs (n,d).
mu : ndarray
mean of the pdf, of shape (k,d). One row of dimension d per
different component, the number of rows k being the number of
component
va : ndarray
variance of the pdf. One row per different component for diagonal
covariance (k, d), or d rows per component for full matrix pdf
(k*d,d).
log : boolean
if True, returns the log-pdf instead of the pdf.
:Returns:
Returns a (n, k) array, each column i being the pdf of the ith mean and
ith variance."""
mu = N.atleast_2d(mu)
va = N.atleast_2d(va)
k = N.shape(mu)[0]
n = N.shape(data)[0]
d = N.shape(mu)[1]
y = N.zeros((k, n))
if N.size(mu) == N.size(va):
for i in range(k):
y[i] = gauss_den(data, mu[i, :], va[i, :], log)
return y.T
else:
for i in range(k):
y[i] = gauss_den(data, mu[i, :], va[d*i:d*i+d, :], log)
return y.T
if __name__ == "__main__":
pass
|