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

[dim_reduce examples] updated plotting of brendan and bgplvm_oil

Browse source
This commit is contained in:
mzwiessele 2014-05-22 15:08:36 +01:00
parent 6534e6606d
commit fa750d4ba4
2 changed files with 13 additions and 13 deletions
Download patch file Download diff file Expand all files Collapse all files

20
GPy/examples/dimensionality_reduction.py
View file

@ -172,14 +172,14 @@ def bgplvm_oil(optimize=True, verbose=1, plot=True, N=200, Q=7, num_inducing=40,
m.data_labels = data['Y'][:N].argmax(axis=1) m.data_labels = data['Y'][:N].argmax(axis=1)
if optimize: if optimize:
m.optimize('scg', messages=verbose, max_iters=max_iters, gtol=.05) m.optimize('bfgs', messages=verbose, max_iters=max_iters, gtol=.05)
if plot: if plot:
fig, (latent_axes, sense_axes) = plt.subplots(1, 2) fig, (latent_axes, sense_axes) = plt.subplots(1, 2)
m.plot_latent(ax=latent_axes, labels=m.data_labels) m.plot_latent(ax=latent_axes, labels=m.data_labels)
data_show = GPy.plotting.matplot_dep.visualize.vector_show((m.Y[0,:])) data_show = GPy.plotting.matplot_dep.visualize.vector_show((m.Y[0,:]))
lvm_visualizer = GPy.plotting.matplot_dep.visualize.lvm_dimselect(param_to_array(m.X.mean)[0:1,:], # @UnusedVariable lvm_visualizer = GPy.plotting.matplot_dep.visualize.lvm_dimselect(param_to_array(m.X.mean)[0:1,:], # @UnusedVariable
m, data_show, latent_axes=latent_axes, sense_axes=sense_axes) m, data_show, latent_axes=latent_axes, sense_axes=sense_axes, labels=m.data_labels)
raw_input('Press enter to finish') raw_input('Press enter to finish')
plt.close(fig) plt.close(fig)
return m return m
@ -386,18 +386,17 @@ def brendan_faces(optimize=True, verbose=True, plot=True):
Yn = Y - Y.mean() Yn = Y - Y.mean()
Yn /= Yn.std() Yn /= Yn.std()
m = GPy.models.GPLVM(Yn, Q) m = GPy.models.BayesianGPLVM(Yn, Q, num_inducing=20)
# optimize # optimize
m.constrain('rbf|noise|white', GPy.transformations.LogexpClipped())
if optimize: m.optimize('scg', messages=verbose, max_iters=1000) if optimize: m.optimize('bfgs', messages=verbose, max_iters=1000)
if plot: if plot:
ax = m.plot_latent(which_indices=(0, 1)) ax = m.plot_latent(which_indices=(0, 1))
y = m.likelihood.Y[0, :] y = m.Y[0, :]
data_show = GPy.plotting.matplot_dep.visualize.image_show(y[None, :], dimensions=(20, 28), transpose=True, order='F', invert=False, scale=False) data_show = GPy.plotting.matplot_dep.visualize.image_show(y[None, :], dimensions=(20, 28), transpose=True, order='F', invert=False, scale=False)
GPy.plotting.matplot_dep.visualize.lvm(m.X[0, :].copy(), m, data_show, ax) lvm = GPy.plotting.matplot_dep.visualize.lvm(m.X.mean[0, :].copy(), m, data_show, ax)
raw_input('Press enter to finish') raw_input('Press enter to finish')
return m return m
@ -411,13 +410,14 @@ def olivetti_faces(optimize=True, verbose=True, plot=True):
Yn = Y - Y.mean() Yn = Y - Y.mean()
Yn /= Yn.std() Yn /= Yn.std()
m = GPy.models.GPLVM(Yn, Q) m = GPy.models.BayesianGPLVM(Yn, Q, num_inducing=20)
if optimize: m.optimize('scg', messages=verbose, max_iters=1000)
if optimize: m.optimize('bfgs', messages=verbose, max_iters=1000)
if plot: if plot:
ax = m.plot_latent(which_indices=(0, 1)) ax = m.plot_latent(which_indices=(0, 1))
y = m.likelihood.Y[0, :] y = m.likelihood.Y[0, :]
data_show = GPy.plotting.matplot_dep.visualize.image_show(y[None, :], dimensions=(112, 92), transpose=False, invert=False, scale=False) data_show = GPy.plotting.matplot_dep.visualize.image_show(y[None, :], dimensions=(112, 92), transpose=False, invert=False, scale=False)
GPy.plotting.matplot_dep.visualize.lvm(m.X[0, :].copy(), m, data_show, ax) lvm = GPy.plotting.matplot_dep.visualize.lvm(m.X.mean[0, :].copy(), m, data_show, ax)
raw_input('Press enter to finish') raw_input('Press enter to finish')
return m return m

6
GPy/plotting/matplot_dep/visualize.py
View file

@ -98,9 +98,9 @@ class lvm(matplotlib_show):
""" """
if vals is None: if vals is None:
if isinstance(model.X, VariationalPosterior): if isinstance(model.X, VariationalPosterior):
vals = param_to_array(model.X.mean) vals = model.X.mean
else: else:
vals = param_to_array(model.X) vals = model.X
if len(vals.shape)==1: if len(vals.shape)==1:
vals = vals[None,:] vals = vals[None,:]
matplotlib_show.__init__(self, vals, axes=latent_axes) matplotlib_show.__init__(self, vals, axes=latent_axes)
@ -133,7 +133,7 @@ class lvm(matplotlib_show):
def modify(self, vals): def modify(self, vals):
"""When latent values are modified update the latent representation and ulso update the output visualization.""" """When latent values are modified update the latent representation and ulso update the output visualization."""
self.vals = vals.copy() self.vals = vals.view(np.ndarray).copy()
y = self.model.predict(self.vals)[0] y = self.model.predict(self.vals)[0]
self.data_visualize.modify(y) self.data_visualize.modify(y)
self.latent_handle.set_data(self.vals[0,self.latent_index[0]], self.vals[0,self.latent_index[1]]) self.latent_handle.set_data(self.vals[0,self.latent_index[0]], self.vals[0,self.latent_index[1]])