diff --git a/GPy/inference/optimization.py b/GPy/inference/optimization.py
index 75cd94ba..1445eed0 100644
--- a/GPy/inference/optimization.py
+++ b/GPy/inference/optimization.py
@@ -29,7 +29,7 @@ class Optimizer():
     :rtype: optimizer object.
 
     """
-    def __init__(self, x_init, messages=False, model = None, max_f_eval=1e4, max_iters = 1e3, ftol=None, gtol=None, xtol=None):
+    def __init__(self, x_init, messages=False, model = None, max_f_eval=1e4, max_iters = 1e3, ftol=None, gtol=None, xtol=None, callback=None):
         self.opt_name = None
         self.x_init = x_init
         self.messages = messages
@@ -45,6 +45,7 @@ class Optimizer():
         self.gtol = gtol
         self.ftol = ftol
         self.model = model
+        self.callback = callback
 
     def run(self, **kwargs):
         start = dt.datetime.now()
@@ -94,6 +95,8 @@ class opt_tnc(Optimizer):
             opt_dict['ftol'] = self.ftol
         if self.gtol is not None:
             opt_dict['pgtol'] = self.gtol
+        if self.callback is not None:
+            opt_dict['callback'] = self.callback
 
         opt_result = optimize.fmin_tnc(f_fp, self.x_init, messages = self.messages,
                        maxfun = self.max_f_eval, **opt_dict)
@@ -128,6 +131,8 @@ class opt_lbfgsb(Optimizer):
             print "WARNING: l-bfgs-b doesn't have an ftol arg, so I'm going to ignore it"
         if self.gtol is not None:
             opt_dict['pgtol'] = self.gtol
+        if self.callback is not None:
+            opt_dict['callback'] = self.callback
 
         opt_result = optimize.fmin_l_bfgs_b(f_fp, self.x_init, iprint = iprint,
                                             maxfun = self.max_f_eval, **opt_dict)
@@ -155,6 +160,8 @@ class opt_simplex(Optimizer):
             opt_dict['ftol'] = self.ftol
         if self.gtol is not None:
             print "WARNING: simplex doesn't have an gtol arg, so I'm going to ignore it"
+        if self.callback is not None:
+            opt_dict['callback'] = self.callback
 
         opt_result = optimize.fmin(f, self.x_init, (), disp = self.messages,
                    maxfun = self.max_f_eval, full_output=True, **opt_dict)
@@ -187,6 +194,8 @@ class opt_rasm(Optimizer):
             print "WARNING: minimize doesn't have an ftol arg, so I'm going to ignore it"
         if self.gtol is not None:
             print "WARNING: minimize doesn't have an gtol arg, so I'm going to ignore it"
+        if self.callback is not None:
+            print "WARNING: minimize doesn't have a callback arg, so I'm going to ignore it"
 
         opt_result = rasm.minimize(self.x_init, f_fp, (), messages = self.messages,
                                    maxnumfuneval = self.max_f_eval)
@@ -205,6 +214,8 @@ class opt_SCG(Optimizer):
     def opt(self, f_fp = None, f = None, fp = None):
         assert not f is None
         assert not fp is None
+        if self.callback is not None:
+            print "WARNING: SCG doesn't have a callback arg, so I'm going to ignore it"
         opt_result = SCG(f,fp,self.x_init, display=self.messages, maxiters=self.max_iters, max_f_eval=self.max_f_eval, xtol=self.xtol, ftol=self.ftol)
         self.x_opt = opt_result[0]
         self.trace = opt_result[1]
diff --git a/GPy/likelihoods/Laplace.py b/GPy/likelihoods/Laplace.py
index 566e4e25..208b1102 100644
--- a/GPy/likelihoods/Laplace.py
+++ b/GPy/likelihoods/Laplace.py
@@ -63,7 +63,6 @@ class Laplace(likelihood):
         return self.likelihood_function._get_param_names()
 
     def _set_params(self, p):
-        print "Setting noise sd: ", p
         return self.likelihood_function._set_params(p)
 
     def both_gradients(self, dL_d_K_Sigma, dK_dthetaK):
diff --git a/GPy/likelihoods/likelihood_functions.py b/GPy/likelihoods/likelihood_functions.py
index 2176aac0..61c79385 100644
--- a/GPy/likelihoods/likelihood_functions.py
+++ b/GPy/likelihoods/likelihood_functions.py
@@ -166,6 +166,8 @@ class student_t(likelihood_function):
         self.log_concave = False
         #super(student_t, self).__init__()
 
+        self._set_params(np.asarray(sigma))
+
     def _get_params(self):
         return np.asarray(self.sigma)
 
@@ -174,6 +176,8 @@ class student_t(likelihood_function):
 
     def _set_params(self, x):
         self.sigma = float(x)
+        print "Setting student t sigma: ", x
+        print x
         #self.covariance_matrix = np.eye(self.N)*self._variance
         #self.precision = 1./self._variance
 
diff --git a/GPy/models/GP.py b/GPy/models/GP.py
index 1682ee6c..79284b59 100644
--- a/GPy/models/GP.py
+++ b/GPy/models/GP.py
@@ -86,6 +86,16 @@ class GP(model):
     def _get_param_names(self):
         return self.kern._get_param_names_transformed() + self.likelihood._get_param_names()
 
+    def _update_params_callback(self, p):
+        #FIXME:Check the transforming
+        #Set the new parameters of the kernel and likelihood within the optimization
+        import ipdb; ipdb.set_trace() ### XXX BREAKPOINT
+        self.kern._set_params_transformed(p[:self.kern.Nparam_transformed()])
+        self.likelihood._set_params(p[self.kern.Nparam_transformed():])
+        #update the likelihood approximation within the optimisation with the current parameters
+        self.update_likelihood_approximation()
+        import ipdb; ipdb.set_trace() ### XXX BREAKPOINT
+
     def update_likelihood_approximation(self):
         """
         Approximates a non-gaussian likelihood using Expectation Propagation