def sample_from( display = True ):
test_x = [ [x] for x in numpy.arrayrange(0,5,.05) ]
test_y = infpy.gp_sample_from( gp, test_x )
if display:
from pylab import figure, plot, show, fill, title
figure()
infpy.gp_plot_prediction( gp.X, test_y )
title('Log likelihood: %f\n%s' % ( gp.log_p_y_given_X, gp.k.params ) )
show()
def predict_values( display = True ):
test_x = [ [x] for x in numpy.arrayrange(0,5,.04) ]
( f_star_mean, V_f_star, log_p_y_given_X ) = gp.predict( test_x )
if display:
from pylab import figure, plot, show, fill, title
figure()
infpy.gp_plot_prediction( test_x, f_star_mean, V_f_star )
plot(
[ x[0] for (x, v) in training_points ],
[ v for (x, v) in training_points ],
'rs' )
infpy.gp_title_and_show( gp )
def gp_ex_fixed_period():
"""Example of the fixed period kernel"""
start, end = -4.0, 0.0
X = infpy.gp_1D_X_range( start, end, 1.3 )
# X = [ ]
# X = [ [ 0.0 ] ]
# X = [ [ 0.0 ], [ 1.0 ] ]
# X = [ [ 0.0 ], [ -1.0 ], [ -2.0 ], [ -3.0 ], ]
y = numpy.asarray( [ math.sin( 2.0 * math.pi * x[0] ) for x in X ] )
# pylab.plot( [ x[0] for x in X ], [ y1 for y1 in y ] )
# pylab.show()
LN = infpy.LogNormalDistribution
k = (
infpy.FixedPeriod1DKernel( 1.0 )
+ infpy.noise_kernel( 0.1 )
)
gp = infpy.GaussianProcess( X, y, k )
sample_X = infpy.gp_1D_X_range( start, end, 0.03 )
y = infpy.gp_sample_from( gp, sample_X )
( y, V_f_star, log_p_y_given_X ) = gp.predict( sample_X )
infpy.gp_plot_prediction( sample_X, y )
infpy.gp_title_and_show( gp )