apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-30 14:35:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

some changes on examples

Browse source
This commit is contained in:
Zhenwen Dai 2014-09-22 15:11:16 +01:00
parent f0df591625
commit ca892e32e1
2 changed files with 13 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

18
GPy/examples/dimensionality_reduction.py
View file

@ -163,7 +163,7 @@ def bgplvm_oil(optimize=True, verbose=1, plot=True, N=200, Q=7, num_inducing=40,
import numpy as np import numpy as np
_np.random.seed(0) _np.random.seed(0)
data = GPy.util.datasets.oil() data = GPy.util.datasets.oil_100()
kernel = GPy.kern.RBF(Q, 1., 1./_np.random.uniform(0,1,(Q,)), ARD=True)# + GPy.kern.Bias(Q, _np.exp(-2)) kernel = GPy.kern.RBF(Q, 1., 1./_np.random.uniform(0,1,(Q,)), ARD=True)# + GPy.kern.Bias(Q, _np.exp(-2))
Y = data['X'][:N] Y = data['X'][:N]
@ -189,12 +189,12 @@ def ssgplvm_oil(optimize=True, verbose=1, plot=True, N=200, Q=7, num_inducing=40
from ..util.misc import param_to_array from ..util.misc import param_to_array
import numpy as np import numpy as np
_np.random.seed(0) #_np.random.seed(0)
data = GPy.util.datasets.oil() data = GPy.util.datasets.oil_100()
kernel = GPy.kern.RBF(Q, 1., 1./_np.random.uniform(0,1,(Q,)), ARD=True)# + GPy.kern.Bias(Q, _np.exp(-2)) kernel = GPy.kern.RBF(Q, 1., 1./_np.random.uniform(0,1,(Q,)), ARD=True)# + GPy.kern.Bias(Q, _np.exp(-2))
Y = data['X'][:N] Y = data['X'][:N]
m = GPy.models.SSGPLVM(Y, Q, kernel=kernel, num_inducing=num_inducing, **k) m = GPy.models.SSGPLVM(Y, Q, kernel=kernel, num_inducing=num_inducing, group_spike=True, **k)
m.data_labels = data['Y'][:N].argmax(axis=1) m.data_labels = data['Y'][:N].argmax(axis=1)
if optimize: if optimize:
@ -327,13 +327,19 @@ def ssgplvm_simulation(optimize=True, verbose=1, group_spike=True,
Y = Ylist[0] Y = Ylist[0]
k = kern.Linear(Q, ARD=True)# + kern.white(Q, _np.exp(-2)) # + kern.bias(Q) k = kern.Linear(Q, ARD=True)# + kern.white(Q, _np.exp(-2)) # + kern.bias(Q)
#k = kern.RBF(Q, ARD=True, lengthscale=10.) #k = kern.RBF(Q, ARD=True, lengthscale=10.)
m = SSGPLVM(Y, Q, init="init", num_inducing=num_inducing, kernel=k, group_spike=group_spike) m = SSGPLVM(Y, Q, init="init", num_inducing=num_inducing, kernel=k, group_spike=group_spike, learnPi=True)
m.X.variance[:] = _np.random.uniform(0,.01,m.X.shape) m.X.variance[:] = _np.random.uniform(0,.01,m.X.shape)
m.likelihood.variance = .1 m.likelihood.variance = .1
if optimize: if optimize:
print "Optimizing model:" print "Optimizing model:"
m.optimize('scg', messages=verbose, max_iters=max_iters, m.likelihood.variance[:] = 0.01
m.likelihood.variance.fix()
m.optimize('bfgs', messages=verbose, max_iters=max_iters,
gtol=.05)
m.likelihood.variance.unfix()
m.likelihood.variance.constrain_positive()
m.optimize('bfgs', messages=verbose, max_iters=max_iters,
gtol=.05) gtol=.05)
if plot: if plot:
m.X.plot("SSGPLVM Latent Space 1D") m.X.plot("SSGPLVM Latent Space 1D")

5
GPy/kern/_src/psi_comp/__init__.py
View file

@ -31,10 +31,7 @@ class PSICOMP_RBF(Pickleable):
def _setup_observers(self): def _setup_observers(self):
pass pass
def _setup_observers(self):
pass
class PSICOMP_Linear(Pickleable): class PSICOMP_Linear(Pickleable):
@Cache_this(limit=2, ignore_args=(0,)) @Cache_this(limit=2, ignore_args=(0,))