00001 # 00002 # Copyright (c) 2009, Georgia Tech Research Corporation 00003 # All rights reserved. 00004 # 00005 # Redistribution and use in source and binary forms, with or without 00006 # modification, are permitted provided that the following conditions are met: 00007 # * Redistributions of source code must retain the above copyright 00008 # notice, this list of conditions and the following disclaimer. 00009 # * Redistributions in binary form must reproduce the above copyright 00010 # notice, this list of conditions and the following disclaimer in the 00011 # documentation and/or other materials provided with the distribution. 00012 # * Neither the name of the Georgia Tech Research Corporation nor the 00013 # names of its contributors may be used to endorse or promote products 00014 # derived from this software without specific prior written permission. 00015 # 00016 # THIS SOFTWARE IS PROVIDED BY GEORGIA TECH RESEARCH CORPORATION ''AS IS'' AND 00017 # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 00018 # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 00019 # DISCLAIMED. IN NO EVENT SHALL GEORGIA TECH BE LIABLE FOR ANY DIRECT, INDIRECT, 00020 # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 00021 # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, 00022 # OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 00023 # LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 00024 # OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 00025 # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 00026 # 00027 00028 # \author Advait Jain (Healthcare Robotics Lab, Georgia Tech.) 00029 00030 00031 # 00032 # standard stuff, but with a bit of visualization. 00033 # 00034 00035 import numpy as np, math 00036 00037 # @return numpy matrix where each column is a new sample. 00038 def sample_gaussian(mean, cov, n_samples): 00039 mean = np.array(mean).flatten() 00040 cov = np.array(cov) 00041 s = np.random.multivariate_normal(mean, cov, (n_samples,)) 00042 return np.matrix(s).T 00043 00044 # x - numpy matrix. each column is a new datapoint 00045 def pca(x): 00046 U, sig, _ = np.linalg.svd(np.matrix(np.cov(x))) 00047 return U, sig 00048 00049 # untested. 00050 def dimen_reduc(x, n_dim): 00051 U, sig = pca(x) 00052 proj_mat = np.matrix(U[:,0:n_dim]) 00053 mn = np.mean(x, 1) 00054 x_projected = proj_mat.T * (x - mn) 00055 return x_projected.A 00056 00057 00058 if __name__ == '__main__': 00059 import matplotlib.pyplot as pp 00060 import roslib; roslib.load_manifest('hrl_lib') 00061 import hrl_lib.matplotlib_util as mpu 00062 00063 pp.axis('equal') 00064 00065 mn = np.matrix((1., 5.)).T 00066 cov = np.matrix([[2.,50.],[50, 100]]) 00067 00068 s = sample_gaussian(mn, cov, 1000) 00069 P = np.cov(s) 00070 mpu.plot_ellipse_cov(mn, P, 'k', 'w') 00071 00072 pp.plot(s[0,:].A1, s[1,:].A1, '.y', ms=3) 00073 00074 U, sig = pca(s) 00075 mn = np.mean(s, 1) 00076 pp.plot(mn[0,:].A1, mn[1,:].A1, 'ok', ms=5) 00077 00078 p1 = mn + U[:,0] * math.sqrt(sig[0]) 00079 p2 = mn + U[:,1] * math.sqrt(sig[1]) 00080 pp.plot([p1[0,0], mn[0,0], p2[0,0]], [p1[1,0], mn[1,0], p2[1,0]], 00081 'k-', linewidth=2) 00082 00083 pp.show() 00084 00085 00086 00087