diff --git a/pysegcnn/core/graphics.py b/pysegcnn/core/graphics.py
index a1b64a9b847d952db6771f2a43bbfffe11e793f4..0b76ac3754719a5a67da60d95b72c0363abc672b 100644
--- a/pysegcnn/core/graphics.py
+++ b/pysegcnn/core/graphics.py
@@ -304,7 +304,7 @@ def plot_sample(x, use_bands, labels,
         # check whether the ground truth is specified and calculate accuracy
         if y is not None and accuracy:
             acc = accuracy_function(v, y)
-            k += ' ({:.2f}%)'.format(acc * 100)
+            k = '{:.2f}%'.format(acc * 100)
 
         # plot model prediction
         ax.imshow(v, cmap=cmap, interpolation='nearest', norm=norm)
diff --git a/pysegcnn/core/trainer.py b/pysegcnn/core/trainer.py
index 689f9c5306929721bf3e2c59f582ee5de4c0694b..206e780c149c9a19a0875b746091d5700dc4db6b 100644
--- a/pysegcnn/core/trainer.py
+++ b/pysegcnn/core/trainer.py
@@ -2591,19 +2591,12 @@ class NetworkInference(BaseConfig):
 
                     # plot current scene
                     if self.plot:
-
-                        # title for prediction
-                        title = ''.join([
-                            (v[0] + str(k)) for k, v in
-                            self.src_ds.dataset.sensor.band_dict().items()
-                            if v in self.bands])
-
                         # plot inputs, ground truth and model predictions
                         fig = plot_sample(inputs.clip(0, 1),
                                           self.bands,
                                           self.use_labels,
                                           y=labels,
-                                          y_pred={title: prdctn},
+                                          y_pred={'Prediction': prdctn},
                                           accuracy=True,
                                           **self.plot_kwargs)