Merge branch 'devel' of github.com:SheffieldML/GPy into devel

GPy/examples/dimensionality_reduction.py

@@ -298,7 +298,7 @@ def mrd_simulation(optimize=True, plot=True, plot_sim=True, **kw):
 
     if optimize:
         print "Optimizing Model:"
-        m.optimize('scg', messages=1, max_iters=5e4, max_f_eval=5e4, gtol=.05)
+        m.optimize('scg', messages=1, max_iters=1e3, max_f_eval=1e3, gtol=.1)
     if plot:
         m.plot_X_1d("MRD Latent Space 1D")
         m.plot_scales("MRD Scales")

GPy/examples/non_Gaussian.py

@@ -1,47 +0,0 @@
-# Copyright (c) 2012, GPy authors (see AUTHORS.txt).
-# Licensed under the BSD 3-clause license (see LICENSE.txt)
-
-
-"""
-Gaussian Processes + Expectation Propagation - Poisson Likelihood
-"""
-import pylab as pb
-import numpy as np
-import GPy
-
-default_seed=10000
-
-def  toy_poisson_1d(seed=default_seed):
-    """
-    Simple 1D classification example
-    :param seed : seed value for data generation (default is 4).
-    :type seed: int
-    """
-
-    X = np.arange(0,100,5)[:,None]
-    F = np.round(np.sin(X/18.) + .1*X) + np.arange(5,25)[:,None]
-    E = np.random.randint(-5,5,20)[:,None]
-    Y = F + E
-
-    kernel = GPy.kern.rbf(1)
-    distribution = GPy.likelihoods.likelihood_functions.Poisson()
-    likelihood = GPy.likelihoods.EP(Y,distribution)
-
-    m = GPy.models.GP(X,likelihood,kernel)
-    m.ensure_default_constraints()
-
-    # Approximate likelihood
-    m.update_likelihood_approximation()
-
-    # Optimize and plot
-    m.optimize()
-    #m.EPEM FIXME
-    print m
-
-    # Plot
-    pb.subplot(211)
-    m.plot_f() #GP plot
-    pb.subplot(212)
-    m.plot() #Output plot
-
-    return m

GPy/likelihoods/likelihood_functions.py

@@ -21,7 +21,7 @@ class LikelihoodFunction(object):
         if link == self._analytical:
             self.moments_match = self._moments_match_analytical
         else:
-            assert isinstance(link,link_functions.link_function)
+            assert isinstance(link,link_functions.LinkFunction)
             self.link = link
             self.moments_match = self._moments_match_numerical
 
@@ -79,7 +79,7 @@ class Binomial(LikelihoodFunction):
     $$
     """
     def __init__(self,link=None):
-        self._analytical = link_functions.probit
+        self._analytical = link_functions.Probit
         if not link:
             link = self._analytical
         super(Binomial, self).__init__(link)
@@ -146,7 +146,7 @@ class Poisson(LikelihoodFunction):
     def __init__(self,link=None):
         self._analytical = None
         if not link:
-            link = link_functions.log()
+            link = link_functions.Log()
         super(Poisson, self).__init__(link)
 
     def _distribution(self,gp,obs):

GPy/likelihoods/link_functions.py

@@ -9,7 +9,7 @@ import pylab as pb
 from ..util.plot import gpplot
 from ..util.univariate_Gaussian import std_norm_pdf,std_norm_cdf
 
-class link_function(object):
+class LinkFunction(object):
     """
     Link function class for doing non-Gaussian likelihoods approximation
 
@@ -19,7 +19,7 @@ class link_function(object):
     def __init__(self):
         pass
 
-class identity(link_function):
+class Identity(LinkFunction):
     def transf(self,mu):
         return mu
 
@@ -29,7 +29,7 @@ class identity(link_function):
     def log_inv_transf(self,f):
         return np.log(f)
 
-class log(link_function):
+class Log(LinkFunction):
 
     def transf(self,mu):
         return np.log(mu)
@@ -40,7 +40,7 @@ class log(link_function):
     def log_inv_transf(self,f):
         return f
 
-class log_ex_1(link_function):
+class Log_ex_1(LinkFunction):
     def transf(self,mu):
         return np.log(np.exp(mu) - 1)
 
@@ -50,11 +50,10 @@ class log_ex_1(link_function):
     def log_inv_tranf(self,f):
         return np.log(np.log(np.exp(f)+1))
 
-class probit(link_function):
+class Probit(LinkFunction):
 
     def inv_transf(self,f):
         return std_norm_cdf(f)
 
     def log_inv_transf(self,f):
         return np.log(std_norm_cdf(f))
-

GPy/models/fitc_classification.py

@@ -16,7 +16,7 @@ class FITCClassification(FITC):
 
     :param X: input observations
     :param Y: observed values
-    :param likelihood: a GPy likelihood, defaults to Binomial with probit link_function
+    :param likelihood: a GPy likelihood, defaults to Binomial with probit link function
     :param kernel: a GPy kernel, defaults to rbf+white
     :param normalize_X:  whether to normalize the input data before computing (predictions will be in original scales)
     :type normalize_X: False|True