Fix qgan.py where batch size is greater than data size to raise error (…

…qiskit-community/qiskit-aqua#1115)

* Update qgan.py

Raise error if the number of the training data items is smaller than the batch size, e.g. due to truncation to bounds.

Co-authored-by: Steve Wood <40241007+woodsp-ibm@users.noreply.github.com>

qiskit/aqua/algorithms/distribution_learners/qgan.py

@@ -270,6 +270,9 @@ def _store_params(self, e, d_loss, g_loss, rel_entr):
     def train(self):
         """
         Train the qGAN
+
+        Raises:
+            AquaError: Batch size bigger than the number of items in the truncated data set
         """
         if self._snapshot_dir is not None:
             with open(os.path.join(self._snapshot_dir, 'output.csv'), mode='w') as csv_file:
@@ -278,6 +281,11 @@ def train(self):
                 writer = csv.DictWriter(csv_file, fieldnames=fieldnames)
                 writer.writeheader()
 
+        if len(self._data) < self._batch_size:
+            raise AquaError(
+                'The batch size needs to be less than the '
+                'truncated data size of {}'.format(len(self._data)))
+
         for e in range(self._num_epochs):
             aqua_globals.random.shuffle(self._data)
             index = 0