added sparsegp with missing data

GPy/kern/constructors.py

@@ -136,7 +136,7 @@ def poly(input_dim,variance=1., weight_variance=None,bias_variance=1.,degree=2,
     part = parts.poly.POLY(input_dim,variance,weight_variance,bias_variance,degree,ARD)
     return kern(input_dim, [part])
 
-def white(input_dim,variance=1.):
+def white(input_dim,variance=1.,name='white'):
     """
      Construct a white kernel.
 
@@ -146,7 +146,7 @@ def white(input_dim,variance=1.):
     :type variance: float
 
     """
-    part = parts.white.White(input_dim,variance)
+    part = parts.white.White(input_dim,variance,name=name)
     return kern(input_dim, [part])
 
 def eq_ode1(output_dim, W=None, rank=1,  kappa=None, length_scale=1., decay=None, delay=None):

GPy/kern/parts/linear.py

@@ -60,7 +60,7 @@ class Linear(Kernpart):
         self._K_computations(X, None)
 
     def update_gradients_full(self, dL_dK, X):
-        #self.variances.gradient[:] = 0
+        self.variances.gradient[:] = 0
         self._param_grad_helper(dL_dK, X, None, self.variances.gradient)
     
     def update_gradients_sparse(self, dL_dKmm, dL_dKnm, dL_dKdiag, X, Z):

GPy/kern/parts/white.py

@@ -15,8 +15,8 @@ class White(Kernpart):
     :param variance:
     :type variance: float
     """
-    def __init__(self,input_dim,variance=1.):
-        super(White, self).__init__(input_dim, 'white')
+    def __init__(self,input_dim,variance=1., name='white'):
+        super(White, self).__init__(input_dim, name)
         self.input_dim = input_dim
         self.variance = Param('variance', variance, Logexp())
         self.add_parameters(self.variance)