fix: use iterrows to iterate pandas dataframe (#2011)

itertuples doesn't return named tuples with large number of column(>254)
with python versions < 3.7. See:
https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.itertuples.html
row._asdict() would fail with regular tuples. Switching to iterrows here
to avoid this limitation.

Co-authored-by: Rui Wang Napieralski <[email protected]>
Co-authored-by: Ajay Karpur <[email protected]>

Author: 3 people

src/sagemaker/feature_store/feature_group.py

@@ -179,10 +179,10 @@ def _ingest_single_batch(
             end_index (int): ending position to ingest in this batch.
         """
         logger.info("Started ingesting index %d to %d", start_index, end_index)
-        for row in data_frame[start_index:end_index].itertuples(index=False):
+        for _, row in data_frame[start_index:end_index].iterrows():
             record = [
                 FeatureValue(feature_name=name, value_as_string=str(value))
-                for name, value in row._asdict().items()
+                for name, value in row.items()
             ]
             sagemaker_session.put_record(
                 feature_group_name=feature_group_name, record=[value.to_dict() for value in record]

tests/unit/sagemaker/feature_store/test_feature_store.py

@@ -146,7 +146,7 @@ def test_load_feature_definition_unsupported_types(sagemaker_session_mock):
 @patch("sagemaker.feature_store.feature_group.IngestionManagerPandas")
 def test_ingest(ingestion_manager_init, sagemaker_session_mock):
     feature_group = FeatureGroup(name="MyGroup", sagemaker_session=sagemaker_session_mock)
-    df = pd.DataFrame({"float": pd.Series([2.0], dtype="float64")})
+    df = pd.DataFrame(dict((f"float{i}", pd.Series([2.0], dtype="float64")) for i in range(300)))
 
     mock_ingestion_manager_instance = Mock()
     ingestion_manager_init.return_value = mock_ingestion_manager_instance