apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-06 02:24:17 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Updated sampling and plots to be correct shape, and changed plotting of sampling to be posterior samples p(f*|f), like it used to be, and samples_y to plot samples of p(y*|y)

Browse source
This commit is contained in:
Alan Saul 2015-09-10 11:08:28 +01:00
parent 6acf5dcfac
commit ef778f75d4
5 changed files with 41 additions and 29 deletions
Download patch file Download diff file Expand all files Collapse all files

12
GPy/core/gp.py
View file

@ -535,7 +535,7 @@ class GP(Model):
which_data_ycols='all', fixed_inputs=[],
levels=20, samples=0, fignum=None, ax=None, resolution=None,
plot_raw=False, linecol=None,fillcol=None, Y_metadata=None,
data_symbol='kx', predict_kw=None, plot_training_data=True):
data_symbol='kx', predict_kw=None, plot_training_data=True, samples_y=0, apply_link=False):
"""
Plot the posterior of the GP.
- In one dimension, the function is plotted with a shaded region identifying two standard deviations.
@ -558,7 +558,7 @@ class GP(Model):
:param levels: number of levels to plot in a contour plot.
:param levels: for 2D plotting, the number of contour levels to use is ax is None, create a new figure
:type levels: int
:param samples: the number of a posteriori samples to plot
:param samples: the number of a posteriori samples to plot, p(f*|y)
:type samples: int
:param fignum: figure to plot on.
:type fignum: figure number
@ -574,6 +574,10 @@ class GP(Model):
:type data_symbol: color either as Tango.colorsHex object or character ('r' is red, 'g' is green) alongside marker type, as is standard in matplotlib.
:param plot_training_data: whether or not to plot the training points
:type plot_training_data: boolean
:param samples_y: the number of a posteriori samples to plot, p(y*|y)
:type samples_y: int
:param apply_link: if there is a link function of the likelihood, plot the link(f*) rather than f*, when plotting posterior samples f
:type apply_link: boolean
"""
assert "matplotlib" in sys.modules, "matplotlib package has not been imported."
from ..plotting.matplot_dep import models_plots
@ -587,7 +591,7 @@ class GP(Model):
levels, samples, fignum, ax, resolution,
plot_raw=plot_raw, Y_metadata=Y_metadata,
data_symbol=data_symbol, predict_kw=predict_kw,
plot_training_data=plot_training_data, **kw)
plot_training_data=plot_training_data, samples_y=samples_y, apply_link=apply_link, **kw)
def plot_data(self, which_data_rows='all',
@ -613,7 +617,7 @@ class GP(Model):
:param levels: number of levels to plot in a contour plot.
:param levels: for 2D plotting, the number of contour levels to use is ax is None, create a new figure
:type levels: int
:param samples: the number of a posteriori samples to plot
:param samples: the number of a posteriori samples to plot, p(f*|y)
:type samples: int
:param fignum: figure to plot on.
:type fignum: figure number

2
GPy/likelihoods/exponential.py
View file

@ -124,7 +124,7 @@ class Exponential(Likelihood):
#d3lik_dlink3 = 6*y/(link_f**4) - 2./(link_f**3)
return d3lik_dlink3
def samples(self, gp):
def samples(self, gp, Y_metadata=None):
"""
Returns a set of samples of observations based on a given value of the latent variable.

2
GPy/likelihoods/likelihood.py
View file

@ -622,7 +622,7 @@ class Likelihood(Parameterized):
Nf_samp = 300
Ny_samp = 1
s = np.random.randn(mu.shape[0], Nf_samp)*np.sqrt(var) + mu
ss_y = self.samples(s, Y_metadata, samples=Ny_samp)
ss_y = self.samples(s, Y_metadata)#, samples=Ny_samp)
#ss_y = ss_y.reshape(mu.shape[0], mu.shape[1], Nf_samp*Ny_samp)
pred_quantiles = [np.percentile(ss_y, q, axis=1)[:,None] for q in quantiles]

8
GPy/likelihoods/poisson.py
View file

@ -137,7 +137,7 @@ class Poisson(Likelihood):
"""
return self.gp_link.transf(gp)
def samples(self, gp, Y_metadata=None, samples=1):
def samples(self, gp, Y_metadata=None):
"""
Returns a set of samples of observations based on a given value of the latent variable.
@ -145,5 +145,7 @@ class Poisson(Likelihood):
"""
orig_shape = gp.shape
gp = gp.flatten()
Ysim = np.random.poisson(self.gp_link.transf(gp), [samples, gp.size]).T
return Ysim.reshape(orig_shape+(samples,))
# Ysim = np.random.poisson(self.gp_link.transf(gp), [samples, gp.size]).T
# return Ysim.reshape(orig_shape+(samples,))
Ysim = np.random.poisson(self.gp_link.transf(gp))
return Ysim.reshape(orig_shape)

46
GPy/plotting/matplot_dep/models_plots.py
View file

@ -75,7 +75,7 @@ def plot_fit(model, plot_limits=None, which_data_rows='all',
levels=20, samples=0, fignum=None, ax=None, resolution=None,
plot_raw=False,
linecol=Tango.colorsHex['darkBlue'],fillcol=Tango.colorsHex['lightBlue'], Y_metadata=None, data_symbol='kx',
apply_link=False, samples_f=0, plot_uncertain_inputs=True, predict_kw=None, plot_training_data=True):
apply_link=False, samples_y=0, plot_uncertain_inputs=True, predict_kw=None, plot_training_data=True):
"""
Plot the posterior of the GP.
- In one dimension, the function is plotted with a shaded region identifying two standard deviations.
@ -93,24 +93,30 @@ def plot_fit(model, plot_limits=None, which_data_rows='all',
:type which_data_rows: 'all' or a list of integers
:param fixed_inputs: a list of tuple [(i,v), (i,v)...], specifying that input index i should be set to value v.
:type fixed_inputs: a list of tuples
:param resolution: the number of intervals to sample the GP on. Defaults to 200 in 1D and 50 (a 50x50 grid) in 2D
:type resolution: int
:param levels: number of levels to plot in a contour plot.
:param levels: for 2D plotting, the number of contour levels to use is ax is None, create a new figure
:type levels: int
:param samples: the number of a posteriori samples to plot p(y*|y)
:param samples: the number of a posteriori samples to plot p(f*|y)
:type samples: int
:param fignum: figure to plot on.
:type fignum: figure number
:param ax: axes to plot on.
:type ax: axes handle
:param resolution: the number of intervals to sample the GP on. Defaults to 200 in 1D and 50 (a 50x50 grid) in 2D
:type resolution: int
:param plot_raw: Whether to plot the raw function p(f|y)
:type plot_raw: boolean
:param linecol: color of line to plot.
:type linecol:
:type linecol: hex or color
:param fillcol: color of fill
:param levels: for 2D plotting, the number of contour levels to use is ax is None, create a new figure
:param apply_link: apply the link function if plotting f (default false)
:type fillcol: hex or color
:param apply_link: apply the link function if plotting f (default false), as well as posterior samples if requested
:type apply_link: boolean
:param samples_f: the number of posteriori f samples to plot p(f*|y)
:type samples_f: int
:param samples_y: the number of posteriori f samples to plot p(y*|y)
:type samples_y: int
:param plot_uncertain_inputs: plot the uncertainty of the inputs as error bars if they have uncertainty (BGPLVM etc.)
:type plot_uncertain_inputs: boolean
:param predict_kw: keyword args for _raw_predict and predict functions if required
:type predict_kw: dict
:param plot_training_data: whether or not to plot the training points
:type plot_training_data: boolean
"""
@ -185,17 +191,17 @@ def plot_fit(model, plot_limits=None, which_data_rows='all',
#optionally plot some samples
if samples: #NOTE not tested with fixed_inputs
Ysim = model.posterior_samples(Xgrid, samples, Y_metadata=Y_metadata)
print(Ysim.shape)
print(Xnew.shape)
for yi in Ysim.T:
plots['posterior_samples'] = ax.plot(Xnew, yi[:,None], '#3300FF', linewidth=0.25)
#ax.plot(Xnew, yi[:,None], marker='x', linestyle='--',color=Tango.colorsHex['darkBlue']) #TODO apply this line for discrete outputs.
if samples_f: #NOTE not tested with fixed_inputs
Fsim = model.posterior_samples_f(Xgrid, samples_f)
Fsim = model.posterior_samples_f(Xgrid, samples)
if apply_link:
Fsim = model.likelihood.gp_link.transf(Fsim)
for fi in Fsim.T:
plots['posterior_samples_f'] = ax.plot(Xnew, fi[:,None], Tango.colorsHex['darkBlue'], linewidth=0.25)
plots['posterior_samples'] = ax.plot(Xnew, fi[:,None], '#3300FF', linewidth=0.25)
#ax.plot(Xnew, fi[:,None], marker='x', linestyle='--',color=Tango.colorsHex['darkBlue']) #TODO apply this line for discrete outputs.
if samples_y: #NOTE not tested with fixed_inputs
Ysim = model.posterior_samples(Xgrid, samples_y, Y_metadata=Y_metadata)
for yi in Ysim.T:
plots['posterior_samples_y'] = ax.scatter(Xnew, yi[:,None], s=5, c=Tango.colorsHex['darkBlue'], marker='o', alpha=0.5)
#ax.plot(Xnew, yi[:,None], marker='x', linestyle='--',color=Tango.colorsHex['darkBlue']) #TODO apply this line for discrete outputs.