bug fix (kernel copy) in mrd

GPy/core/transformations.py

@@ -39,8 +39,8 @@ class logexp(transformation):
         return '(+ve)'
 
 class logexp_clipped(transformation):
-    max_bound = 1e250
-    min_bound = 1e-9
+    max_bound = 1e300
+    min_bound = 1e-10
     log_max_bound = np.log(max_bound)
     log_min_bound = np.log(min_bound)
     def __init__(self, lower=1e-6):
@@ -51,7 +51,7 @@ class logexp_clipped(transformation):
         f = np.log(1. + exp)
 #         if np.isnan(f).any():
 #             import ipdb;ipdb.set_trace()
-        return f
+        return np.clip(f, self.min_bound, self.max_bound)
     def finv(self, f):
         return np.log(np.exp(np.clip(f, self.min_bound, self.max_bound)) - 1.)
     def gradfactor(self, f):

GPy/examples/dimensionality_reduction.py

@@ -278,7 +278,7 @@ def bgplvm_simulation(optimize='scg',
     return m
 
 def mrd_simulation(optimize=True, plot_sim=False, **kw):
-    D1, D2, D3, N, M, Q = 15, 8, 8, 100, 3, 7
+    D1, D2, D3, N, M, Q = 150, 200, 400, 700, 3, 7
     slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, M, Q, plot_sim)
 
     from GPy.models import mrd
@@ -288,12 +288,12 @@ def mrd_simulation(optimize=True, plot_sim=False, **kw):
     reload(mrd); reload(kern)
 
     k = kern.linear(Q, [0.01] * Q, True) + kern.bias(Q, np.exp(-2)) + kern.white(Q, np.exp(-2))
-    m = mrd.MRD(Ylist, Q=Q, M=M, kernel=k, initx="concat", initz='permute', **kw)
+    m = mrd.MRD(Ylist, Q=Q, M=M, kernels=k, initx="concat", initz='permute', **kw)
 
     for i, Y in enumerate(Ylist):
         m['{}_noise'.format(i + 1)] = Y.var() / 100.
 
-    m.constrain('variance|noise', logexp_clipped())
+    # m.constrain('variance|noise', logexp_clipped(1e-6))
     m.ensure_default_constraints()
 
     # DEBUG

GPy/models/Bayesian_GPLVM.py

@@ -95,7 +95,7 @@ class Bayesian_GPLVM(sparse_GP, GPLVM):
         return x
 
     def _clipped(self, x):
-        return x # np.clip(x, -1e100, 1e100)
+        return x # np.clip(x, -1e300, 1e300)
     
     def _set_params(self, x, save_old=True, save_count=0):
 #         try:

GPy/models/mrd.py

@@ -44,7 +44,7 @@ class MRD(model):
     """
     def __init__(self, likelihood_or_Y_list, Q, M=10, names=None,
                  kernels=None, initx='PCA',
-                 initz='permute', _debug=False, **kwargs):
+                 initz='permute', _debug=False, **kw):
         if names is None:
             self.names = ["{}".format(i + 1) for i in range(len(likelihood_or_Y_list))]
 
@@ -64,7 +64,7 @@ class MRD(model):
         self._init = True
         X = self._init_X(initx, likelihood_or_Y_list)
         Z = self._init_Z(initz, X)
-        self.bgplvms = [Bayesian_GPLVM(l, k, X=X, Z=Z, M=self.M, **kwargs) for l, k in zip(likelihood_or_Y_list, kernels)]
+        self.bgplvms = [Bayesian_GPLVM(l, Q=Q, kernel=k, X=X, Z=Z, M=self.M, **kw) for l, k in zip(likelihood_or_Y_list, kernels)]
         del self._init
 
         self.gref = self.bgplvms[0]
@@ -229,12 +229,12 @@ class MRD(model):
             else:
                 Ylist.append(likelihood_or_Y.Y)
         del likelihood_list
-        if init in "PCA_single":
+        if init in "PCA_concat":
+            X = PCA(numpy.hstack(Ylist), self.Q)[0]
+        elif init in "PCA_single":
             X = numpy.zeros((Ylist[0].shape[0], self.Q))
             for qs, Y in itertools.izip(numpy.array_split(numpy.arange(self.Q), len(Ylist)), Ylist):
                 X[:, qs] = PCA(Y, len(qs))[0]
-        elif init in "PCA_concat":
-            X = PCA(numpy.hstack(Ylist), self.Q)[0]
         else: # init == 'random':
             X = numpy.random.randn(Ylist[0].shape[0], self.Q)
         self.X = X