Minor fix

examples/v1alpha1/cifar10/configmap_cifar.yaml

@@ -2,7 +2,6 @@ apiVersion: v1
 kind: ConfigMap
 metadata:
   name: dist-train-cifar
-  namespace: kubeflow
 data:
   dist_train_cifar.py: |
     #!/usr/bin/env python
@@ -114,7 +113,7 @@ data:
         partition = DataPartitioner(train_set_full, partition_sizes)
         partition = partition.use(dist.get_rank())
         train_set = torch.utils.data.DataLoader(
-            partition, batch_size=int(bsz), shuffle=True)
+            partition, batch_size=bsz, shuffle=True)
         test_set = torch.utils.data.DataLoader(test_set_full, batch_size=BATCH_SIZE, shuffle=False)
         return train_set, test_set, bsz