apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-10 12:32:40 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

last stability changes

Browse source
This commit is contained in:
Max Zwiessele 2013-05-20 10:11:27 +01:00
parent 3c12f85a28
commit 99017e5e50
4 changed files with 14 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

8
GPy/core/transformations.py
View file

@ -39,8 +39,8 @@ class logexp(transformation):
return '(+ve)' return '(+ve)'
class logexp_clipped(transformation): class logexp_clipped(transformation):
max_bound = 1e300 max_bound = 1e250
min_bound = 1e-10 min_bound = 1e-9
log_max_bound = np.log(max_bound) log_max_bound = np.log(max_bound)
log_min_bound = np.log(min_bound) log_min_bound = np.log(min_bound)
def __init__(self, lower=1e-6): def __init__(self, lower=1e-6):
@ -49,11 +49,13 @@ class logexp_clipped(transformation):
def f(self, x): def f(self, x):
exp = np.exp(np.clip(x, self.log_min_bound, self.log_max_bound)) exp = np.exp(np.clip(x, self.log_min_bound, self.log_max_bound))
f = np.log(1. + exp) f = np.log(1. + exp)
if np.isnan(f).any():
import ipdb;ipdb.set_trace()
return f return f
def finv(self, f): def finv(self, f):
return np.log(np.exp(np.clip(f, self.min_bound, self.max_bound)) - 1.) return np.log(np.exp(np.clip(f, self.min_bound, self.max_bound)) - 1.)
def gradfactor(self, f): def gradfactor(self, f):
ef = np.exp(f) ef = np.exp(f) # np.clip(f, self.min_bound, self.max_bound))
gf = (ef - 1.) / ef gf = (ef - 1.) / ef
return np.where(f < self.lower, 0, gf) return np.where(f < self.lower, 0, gf)
def initialize(self, f): def initialize(self, f):

11
GPy/examples/dimensionality_reduction.py
View file

@ -273,8 +273,8 @@ def bgplvm_simulation(optimize='scg',
pylab.figure(); pylab.axis(); m.kern.plot_ARD() pylab.figure(); pylab.axis(); m.kern.plot_ARD()
return m return m
def mrd_simulation(plot_sim=False): def mrd_simulation(optimize=True, plot_sim=False):
D1, D2, D3, N, M, Q = 150, 250, 300, 700, 3, 7 D1, D2, D3, N, M, Q = 150, 250, 30, 300, 3, 7
slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, M, Q, plot_sim) slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, M, Q, plot_sim)
from GPy.models import mrd from GPy.models import mrd
@ -292,6 +292,13 @@ def mrd_simulation(plot_sim=False):
m.constrain('variance|noise', logexp_clipped()) m.constrain('variance|noise', logexp_clipped())
m.ensure_default_constraints() m.ensure_default_constraints()
# DEBUG
np.seterr("raise")
if optimize:
print "Optimizing Model:"
m.optimize('scg', messages=1, max_iters=3e3)
return m return m
def brendan_faces(): def brendan_faces():

2
GPy/inference/SCG.py
View file

@ -85,8 +85,6 @@ def SCG(f, gradf, x, optargs=(), maxiters=500, max_f_eval=500, display=True, xto
# Increase effective curvature and evaluate step size alpha. # Increase effective curvature and evaluate step size alpha.
delta = theta + beta * kappa delta = theta + beta * kappa
if delta <= 0: if delta <= 0:
if display:
print ""
delta = beta * kappa delta = beta * kappa
beta = beta - theta / kappa beta = beta - theta / kappa

3
GPy/models/Bayesian_GPLVM.py
View file

@ -171,9 +171,6 @@ class Bayesian_GPLVM(sparse_GP, GPLVM):
self.dbound_dZtheta = sparse_GP._log_likelihood_gradients(self) self.dbound_dZtheta = sparse_GP._log_likelihood_gradients(self)
return np.hstack((self.dbound_dmuS.flatten(), self.dbound_dZtheta)) return np.hstack((self.dbound_dmuS.flatten(), self.dbound_dZtheta))
def _log_likelihood_normal_gradients(self):
Si, _, _, _ = pdinv(self.X_variance)
def plot_latent(self, which_indices=None, *args, **kwargs): def plot_latent(self, which_indices=None, *args, **kwargs):
if which_indices is None: if which_indices is None: