diff --git a/pts/model/transformer/transformer_estimator.py b/pts/model/transformer/transformer_estimator.py
index feaed5a..a9cbb5b 100644
--- a/pts/model/transformer/transformer_estimator.py
+++ b/pts/model/transformer/transformer_estimator.py
@@ -105,12 +105,13 @@ class TransformerEstimator(PTSEstimator):
         return Chain(
             [RemoveFields(field_names=remove_field_names)]
             + (
-                [SetField(output_field=FieldName.FEAT_STATIC_CAT, value=[0.0])]
+                [SetField(output_field=FieldName.FEAT_STATIC_CAT, value=[0])]
                 if not self.use_feat_static_cat
                 else []
             )
             + [
-                AsNumpyArray(field=FieldName.FEAT_STATIC_CAT, expected_ndim=1),
+                AsNumpyArray(field=FieldName.FEAT_STATIC_CAT,
+                             expected_ndim=1, dtype=np.long),
                 AsNumpyArray(
                     field=FieldName.TARGET,
                     # in the following line, we add 1 for the time dimension
diff --git a/pts/model/transformer/transformer_network.py b/pts/model/transformer/transformer_network.py
index b3f7585..c5650d8 100644
--- a/pts/model/transformer/transformer_network.py
+++ b/pts/model/transformer/transformer_network.py
@@ -70,13 +70,13 @@ class TransformerNetwork(nn.Module):
 
         self.embedder = FeatureEmbedder(
             cardinalities=cardinality,
-            embedding_dims=[embedding_dimension for _ in cardinality],
+            embedding_dims=embedding_dimension,
         )
 
         if scaling:
-            self.scaler = MeanScaler(keepdims=True)
+            self.scaler = MeanScaler(keepdim=True)
         else:
-            self.scaler = NOPScaler(keepdims=True)
+            self.scaler = NOPScaler(keepdim=True)
 
     @staticmethod
     def get_lagged_subsequences(
@@ -255,10 +255,10 @@ class TransformerTrainingNetwork(TransformerNetwork):
             future_target=future_target,
         )
 
-        enc_input = input[:, :self.context_length, ...]  # F.slice_axis(
+        enc_input = inputs[:, :self.context_length, ...]  # F.slice_axis(
         #     inputs, axis=1, begin=0, end=self.context_length
         # )
-        dec_input = input[:, self.context_length:, ...]  # F.slice_axis(
+        dec_input = inputs[:, self.context_length:, ...]  # F.slice_axis(
         #     inputs, axis=1, begin=self.context_length, end=None
         # )
 
diff --git a/runs/Jan27_20-21-12_DL-Box/events.out.tfevents.1580152872.DL-Box.11613.0 b/runs/Jan27_20-21-12_DL-Box/events.out.tfevents.1580152872.DL-Box.11613.0
new file mode 100644
index 0000000..eb649aa
Binary files /dev/null and b/runs/Jan27_20-21-12_DL-Box/events.out.tfevents.1580152872.DL-Box.11613.0 differ
diff --git a/runs/Jan27_20-23-51_DL-Box/events.out.tfevents.1580153031.DL-Box.11858.0 b/runs/Jan27_20-23-51_DL-Box/events.out.tfevents.1580153031.DL-Box.11858.0
new file mode 100644
index 0000000..5283630
Binary files /dev/null and b/runs/Jan27_20-23-51_DL-Box/events.out.tfevents.1580153031.DL-Box.11858.0 differ
diff --git a/runs/Jan27_20-24-54_DL-Box/events.out.tfevents.1580153094.DL-Box.11858.1 b/runs/Jan27_20-24-54_DL-Box/events.out.tfevents.1580153094.DL-Box.11858.1
new file mode 100644
index 0000000..5dba200
Binary files /dev/null and b/runs/Jan27_20-24-54_DL-Box/events.out.tfevents.1580153094.DL-Box.11858.1 differ
diff --git a/runs/Jan27_20-26-28_DL-Box/events.out.tfevents.1580153188.DL-Box.12203.0 b/runs/Jan27_20-26-28_DL-Box/events.out.tfevents.1580153188.DL-Box.12203.0
new file mode 100644
index 0000000..44faf35
Binary files /dev/null and b/runs/Jan27_20-26-28_DL-Box/events.out.tfevents.1580153188.DL-Box.12203.0 differ
diff --git a/runs/Jan27_20-27-31_DL-Box/events.out.tfevents.1580153251.DL-Box.12379.0 b/runs/Jan27_20-27-31_DL-Box/events.out.tfevents.1580153251.DL-Box.12379.0
new file mode 100644
index 0000000..23fa7b0
Binary files /dev/null and b/runs/Jan27_20-27-31_DL-Box/events.out.tfevents.1580153251.DL-Box.12379.0 differ