rm expensive lines in estimator related to chain log

twin4build/estimator/estimator.py

@@ -210,7 +210,8 @@ def run_emcee_estimation(self,
             x0_start = x0_start.reshape((n_temperature, n_walkers, ndim))
 
         elif walker_initialization=="hypercube":
-            x0_start = np.random.uniform(low=self.x0-1e-5, high=self.x0+1e-5, size=(n_temperature, n_walkers, ndim))
+            r = 1e-5
+            x0_start = np.random.uniform(low=self.x0-r, high=self.x0+r, size=(n_temperature, n_walkers, ndim))
             lb = np.resize(self.lb,(x0_start.shape))
             ub = np.resize(self.ub,(x0_start.shape))
             x0_start[x0_start<self.lb] = lb[x0_start<self.lb]

twin4build/simulator/simulator.py

@@ -380,7 +380,7 @@ def _sim_func_gaussian_process(self, model, theta, stepSize, startTime, endTime,
         except FMICallException as inst:
             return None
 
-        return (y, y_noise, y_model)
+        return (y, y_model, y_noise)
 
     def _sim_func_wrapped(self, args):
             return self._sim_func(*args)
@@ -445,8 +445,10 @@ def run_emcee_inference(self, model, parameter_chain, targetParameters, targetMe
             # standardDeviation = np.array([el["standardDeviation"] for el in targetMeasuringDevices.values()])
             # y_w_obs_error = y# + np.random.normal(0, standardDeviation, size=y.shape)
             for col in range(len(targetMeasuringDevices)):
-                predictions_noise[col].append(y[1][:,col])
-                predictions_model[col].append(y[2][:,col])
+                if assume_uncorrelated_noise==False:
+                    predictions_noise[col].append(y[2][:,col])
+
+                predictions_model[col].append(y[1][:,col])
                 predictions[col].append(y[0][:,col])
 
         intervals = []