diff --git a/brian2modelfitting/inferencer.py b/brian2modelfitting/inferencer.py
index 60f17a2..fadc6ae 100644
--- a/brian2modelfitting/inferencer.py
+++ b/brian2modelfitting/inferencer.py
@@ -1101,8 +1101,8 @@ def pairplot(self, samples=None, points=None, limits=None, subset=None,
                 if len(lim_vals) != 2:
                     raise ValueError('Invalid limits for parameter: '
                                      f'{param_name}')
-            limits = [[limits[param_name][0].item(),
-                       limits[param_name][1].item()]
+            limits = [[float(limits[param_name][0]),
+                       float(limits[param_name][1])]
                       for param_name in self.param_names]
         if subset:
             for param_name in subset:
@@ -1117,8 +1117,8 @@ def pairplot(self, samples=None, points=None, limits=None, subset=None,
                 if len(lim_vals) != 2:
                     raise ValueError('Invalid limits for parameter: '
                                      f'{param_name}')
-            ticks = [[ticks[param_name][0].item(),
-                      ticks[param_name][1].item()]
+            ticks = [[float(ticks[param_name][0]),
+                      float(ticks[param_name][1])]
                      for param_name in self.param_names]
         else:
             ticks = []
@@ -1207,12 +1207,12 @@ def conditional_pairplot(self, condition, density=None, points=None,
                 if len(lim_vals) != 2:
                     raise ValueError('Invalid limits for parameter: '
                                      f'{param_name}')
-            limits = [[limits[param_name][0].item(),
-                       limits[param_name][1].item()]
+            limits = [[float(limits[param_name][0]),
+                       float(limits[param_name][1])]
                       for param_name in self.param_names]
         else:
-            limits = [[self.params[param_name][0].item(),
-                       self.params[param_name][1].item()]
+            limits = [[float(self.params[param_name][0]),
+                       float(self.params[param_name][1])]
                       for param_name in self.param_names]
         if subset:
             for param_name in subset:
@@ -1227,8 +1227,8 @@ def conditional_pairplot(self, condition, density=None, points=None,
                 if len(lim_vals) != 2:
                     raise ValueError('Invalid limits for parameter: '
                                      f'{param_name}')
-            ticks = [[ticks[param_name][0].item(),
-                      ticks[param_name][1].item()]
+            ticks = [[float(ticks[param_name][0]),
+                      float(ticks[param_name][1])]
                      for param_name in self.param_names]
         else:
             ticks = []
@@ -1237,14 +1237,15 @@ def conditional_pairplot(self, condition, density=None, points=None,
                 if param_name not in self.param_names:
                     raise AttributeError(f'Invalid parameter: {param_name}')
             labels = [labels[param_name] for param_name in self.param_names]
-        fig, axes = sbi.analysis.conditional_pairplot(density=d,
-                                                      condition=condition,
-                                                      limits=limits,
-                                                      points=points,
-                                                      subset=subset,
-                                                      labels=labels,
-                                                      ticks=ticks,
-                                                      **kwargs)
+        from sbi.analysis import conditional_pairplot
+        fig, axes = conditional_pairplot(density=d,
+                                         condition=condition,
+                                         limits=limits,
+                                         points=points,
+                                         subset=subset,
+                                         labels=labels,
+                                         ticks=ticks,
+                                         **kwargs)
         return fig, axes
 
     def conditional_corrcoeff(self, condition, density=None, limits=None,
@@ -1305,12 +1306,12 @@ def conditional_corrcoeff(self, condition, density=None, limits=None,
                 if len(lim_vals) != 2:
                     raise ValueError('Invalid limits for parameter: '
                                      f'{param_name}')
-            limits = [[limits[param_name][0].item(),
-                       limits[param_name][1].item()]
+            limits = [[float(limits[param_name][0]),
+                       float(limits[param_name][1])]
                       for param_name in self.param_names]
         else:
-            limits = [[self.params[param_name][0].item(),
-                       self.params[param_name][1].item()]
+            limits = [[float(self.params[param_name][0]),
+                       float(self.params[param_name][1])]
                       for param_name in self.param_names]
         limits = torch.tensor(limits)
         if subset:
@@ -1319,11 +1320,12 @@ def conditional_corrcoeff(self, condition, density=None, limits=None,
                     raise AttributeError(f'Invalid parameter: {param_name}')
             subset = [self.param_names.index(param_name)
                       for param_name in subset]
-        cond_coeff = sbi.analysis.conditional_corrcoeff(density=d,
-                                                        limits=limits,
-                                                        condition=condition,
-                                                        subset=subset,
-                                                        **kwargs)
+        from sbi.analysis import conditional_corrcoeff
+        cond_coeff = conditional_corrcoeff(density=d,
+                                           limits=limits,
+                                           condition=condition,
+                                           subset=subset,
+                                           **kwargs)
         return cond_coeff.numpy()
 
     def generate_traces(self, n_samples=1, posterior=None, output_var=None,