apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-04-28 06:16:24 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

optional plotting of inducing inputs added

Browse source
This commit is contained in:
Max Zwiessele 2013-06-28 11:01:31 +01:00
parent 7325e319b4
commit c0d514b6c0
1 changed files with 24 additions and 35 deletions
Download patch file Download diff file Expand all files Collapse all files

59
GPy/util/plot_latent.py
View file

@ -2,7 +2,7 @@ import pylab as pb
import numpy as np import numpy as np
from .. import util from .. import util
def plot_latent(model, labels=None, which_indices=None, resolution=50, ax=None, marker='o', s=40, fignum=None): def plot_latent(model, labels=None, which_indices=None, resolution=50, ax=None, marker='o', s=40, fignum=None, plot_inducing=False, legend=True):
""" """
:param labels: a np.array of size model.num_data containing labels for the points (can be number, strings, etc) :param labels: a np.array of size model.num_data containing labels for the points (can be number, strings, etc)
:param resolution: the resolution of the grid on which to evaluate the predictive variance :param resolution: the resolution of the grid on which to evaluate the predictive variance
@ -15,11 +15,11 @@ def plot_latent(model, labels=None, which_indices=None, resolution=50, ax=None,
if labels is None: if labels is None:
labels = np.ones(model.num_data) labels = np.ones(model.num_data)
if which_indices is None: if which_indices is None:
if model.input_dim==1: if model.input_dim == 1:
input_1 = 0 input_1 = 0
input_2 = None input_2 = None
if model.input_dim==2: if model.input_dim == 2:
input_1, input_2 = 0,1 input_1, input_2 = 0, 1
else: else:
try: try:
input_1, input_2 = np.argsort(model.input_sensitivity())[:2] input_1, input_2 = np.argsort(model.input_sensitivity())[:2]
@ -28,14 +28,14 @@ def plot_latent(model, labels=None, which_indices=None, resolution=50, ax=None,
else: else:
input_1, input_2 = which_indices input_1, input_2 = which_indices
#first, plot the output variance as a function of the latent space # first, plot the output variance as a function of the latent space
Xtest, xx,yy,xmin,xmax = util.plot.x_frame2D(model.X[:,[input_1, input_2]],resolution=resolution) Xtest, xx, yy, xmin, xmax = util.plot.x_frame2D(model.X[:, [input_1, input_2]], resolution=resolution)
Xtest_full = np.zeros((Xtest.shape[0], model.X.shape[1])) Xtest_full = np.zeros((Xtest.shape[0], model.X.shape[1]))
Xtest_full[:, :2] = Xtest Xtest_full[:, :2] = Xtest
mu, var, low, up = model.predict(Xtest_full) mu, var, low, up = model.predict(Xtest_full)
var = var[:, :1] var = var[:, :1]
ax.imshow(var.reshape(resolution, resolution).T, ax.imshow(var.reshape(resolution, resolution).T,
extent=[xmin[0], xmax[0], xmin[1], xmax[1]], cmap=pb.cm.binary,interpolation='bilinear',origin='lower') extent=[xmin[0], xmax[0], xmin[1], xmax[1]], cmap=pb.cm.binary, interpolation='bilinear', origin='lower')
# make sure labels are in order of input: # make sure labels are in order of input:
ulabels = [] ulabels = []
@ -47,46 +47,35 @@ def plot_latent(model, labels=None, which_indices=None, resolution=50, ax=None,
if type(ul) is np.string_: if type(ul) is np.string_:
this_label = ul this_label = ul
elif type(ul) is np.int64: elif type(ul) is np.int64:
this_label = 'class %i'%ul this_label = 'class %i' % ul
else: else:
this_label = 'class %i'%i this_label = 'class %i' % i
if len(marker) == len(ulabels): if len(marker) == len(ulabels):
m = marker[i] m = marker[i]
else: else:
m = marker m = marker
index = np.nonzero(labels==ul)[0] index = np.nonzero(labels == ul)[0]
if model.input_dim==1: if model.input_dim == 1:
x = model.X[index,input_1] x = model.X[index, input_1]
y = np.zeros(index.size) y = np.zeros(index.size)
else: else:
x = model.X[index,input_1] x = model.X[index, input_1]
y = model.X[index,input_2] y = model.X[index, input_2]
ax.scatter(x, y, marker=m, s=s, color=util.plot.Tango.nextMedium(), label=this_label) ax.scatter(x, y, marker=m, s=s, color=util.plot.Tango.nextMedium(), label=this_label)
ax.set_xlabel('latent dimension %i'%input_1) ax.set_xlabel('latent dimension %i' % input_1)
ax.set_ylabel('latent dimension %i'%input_2) ax.set_ylabel('latent dimension %i' % input_2)
if not np.all(labels==1.): if not np.all(labels == 1.) and legend:
ax.legend(loc=0,numpoints=1) ax.legend(loc=0, numpoints=1)
ax.set_xlim(xmin[0],xmax[0]) ax.set_xlim(xmin[0], xmax[0])
ax.set_ylim(xmin[1],xmax[1]) ax.set_ylim(xmin[1], xmax[1])
ax.grid(b=False) # remove the grid if present, it doesn't look good ax.grid(b=False) # remove the grid if present, it doesn't look good
ax.set_aspect('auto') # set a nice aspect ratio ax.set_aspect('auto') # set a nice aspect ratio
return ax
if plot_inducing:
ax.plot(model.Z[:, input_1], model.Z[:, input_2], '^w')
def plot_latent_indices(Model, which_indices=None, *args, **kwargs):
if which_indices is None:
try:
input_1, input_2 = np.argsort(Model.input_sensitivity())[:2]
except:
raise ValueError, "cannot Automatically determine which dimensions to plot, please pass 'which_indices'"
else:
input_1, input_2 = which_indices
ax = plot_latent(Model, which_indices=[input_1, input_2], *args, **kwargs)
# TODO: Here test if there are inducing points...
ax.plot(Model.Z[:, input_1], Model.Z[:, input_2], '^w')
return ax return ax