apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-10 20:42:39 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

[pca] pca -> PCA

Browse source
This commit is contained in:
Max Zwiessele 2014-10-17 11:45:04 +01:00
parent e12329c626
commit 5ac7b2c524
2 changed files with 10 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

4
GPy/util/initialization.py
View file

@ -5,12 +5,12 @@ Created on 24 Feb 2014
''' '''
import numpy as np import numpy as np
from GPy.util.pca import pca from GPy.util.pca import PCA
def initialize_latent(init, input_dim, Y): def initialize_latent(init, input_dim, Y):
Xr = np.asfortranarray(np.random.randn(Y.shape[0], input_dim)) Xr = np.asfortranarray(np.random.randn(Y.shape[0], input_dim))
if init == 'PCA': if init == 'PCA':
p = pca(Y) p = PCA(Y)
PC = p.project(Y, min(input_dim, Y.shape[1])) PC = p.project(Y, min(input_dim, Y.shape[1]))
Xr[:PC.shape[0], :PC.shape[1]] = PC Xr[:PC.shape[0], :PC.shape[1]] = PC
var = .1*p.fracs[:input_dim] var = .1*p.fracs[:input_dim]

13
GPy/util/pca.py
View file

@ -14,9 +14,9 @@ from numpy.linalg.linalg import LinAlgError
from operator import setitem from operator import setitem
import itertools import itertools
class pca(object): class PCA(object):
""" """
pca module with automatic primal/dual determination. PCA module with automatic primal/dual determination.
""" """
def __init__(self, X): def __init__(self, X):
self.mu = None self.mu = None
@ -39,7 +39,7 @@ class pca(object):
def center(self, X): def center(self, X):
""" """
Center `X` in pca space. Center `X` in PCA space.
""" """
X = X.copy() X = X.copy()
inan = numpy.isnan(X) inan = numpy.isnan(X)
@ -66,7 +66,7 @@ class pca(object):
def project(self, X, Q=None): def project(self, X, Q=None):
""" """
Project X into pca space, defined by the Q highest eigenvalues. Project X into PCA space, defined by the Q highest eigenvalues.
Y = X dot V Y = X dot V
""" """
if Q is None: if Q is None:
@ -80,13 +80,16 @@ class pca(object):
""" """
Plot fractions of Eigenvalues sorted in descending order. Plot fractions of Eigenvalues sorted in descending order.
""" """
from GPy.plotting.matplot_dep import Tango
Tango.reset()
col = Tango.nextMedium()
if ax is None: if ax is None:
fig = pylab.figure(fignum) fig = pylab.figure(fignum)
ax = fig.add_subplot(111) ax = fig.add_subplot(111)
if Q is None: if Q is None:
Q = self.Q Q = self.Q
ticks = numpy.arange(Q) ticks = numpy.arange(Q)
bar = ax.bar(ticks - .4, self.fracs[:Q]) bar = ax.bar(ticks - .4, self.fracs[:Q], color=col)
ax.set_xticks(ticks, map(lambda x: r"${}$".format(x), ticks + 1)) ax.set_xticks(ticks, map(lambda x: r"${}$".format(x), ticks + 1))
ax.set_ylabel("Eigenvalue fraction") ax.set_ylabel("Eigenvalue fraction")
ax.set_xlabel("PC") ax.set_xlabel("PC")