Merge pull request #25 from philschmid/add-error-for-csv-without-header

Add exception for CSV without headers.

Author: philschmid

src/sagemaker_huggingface_inference_toolkit/decoder_encoder.py

@@ -11,22 +11,17 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import csv
 import datetime
 import json
 from io import StringIO
-import csv
+
 import numpy as np
-from sagemaker_inference.decoder import (
-    _npy_to_numpy,
-    _npz_to_sparse,
-)
-from sagemaker_inference.encoder import (
-    _array_to_npy,
-)
-from sagemaker_inference import (
-    content_types,
-    errors,
-)
+from sagemaker_inference import content_types, errors
+from sagemaker_inference.decoder import _npy_to_numpy, _npz_to_sparse
+from sagemaker_inference.encoder import _array_to_npy
+
+from mms.service import PredictionException
 
 
 def decode_json(content):
@@ -42,6 +37,13 @@ def decode_csv(string_like):  # type: (str) -> np.array
         (dict): dictonatry for input
     """
     stream = StringIO(string_like)
+    # detects if the incoming csv has headers
+    if not any(header in string_like.splitlines()[0].lower() for header in ["question", "context", "inputs"]):
+        raise PredictionException(
+            f"You need to provide the correct CSV with Header columns to use it with the inference toolkit default handler.",
+            400,
+        )
+    # reads csv as io
     request_list = list(csv.DictReader(stream))
     if "inputs" in request_list[0].keys():
         return {"inputs": [entry["inputs"] for entry in request_list]}
@@ -123,6 +125,8 @@ def decode(content, content_type=content_types.JSON):
         return decoder(content)
     except KeyError:
         raise errors.UnsupportedFormatError(content_type)
+    except PredictionException as pred_err:
+        raise pred_err
 
 
 def encode(content, content_type=content_types.JSON):

src/sagemaker_huggingface_inference_toolkit/handler_service.py

@@ -15,15 +15,15 @@
 import importlib
 import logging
 import os
-import time
 import sys
+import time
 from abc import ABC
 
-from sagemaker_inference import environment, utils, content_types
+from sagemaker_inference import content_types, environment, utils
 from transformers.pipelines import SUPPORTED_TASKS
 
-from mms.service import PredictionException
 from mms import metrics
+from mms.service import PredictionException
 from sagemaker_huggingface_inference_toolkit import decoder_encoder
 from sagemaker_huggingface_inference_toolkit.transformers_utils import (
     _is_gpu_available,

tests/unit/test_decoder_encoder.py

@@ -13,6 +13,9 @@
 # limitations under the License.
 import json
 
+import pytest
+
+from mms.service import PredictionException
 from sagemaker_huggingface_inference_toolkit import decoder_encoder
 
 
@@ -46,6 +49,13 @@ def test_decode_csv():
     assert decoded_data == {"inputs": ["I love you", "I like you"]}
 
 
+def test_decode_csv_without_header():
+    with pytest.raises(PredictionException):
+        decoder_encoder.decode_csv(
+            "where do i live?,My name is Philipp and I live in Nuremberg\r\nwhere is Berlin?,Berlin is the capital of Germany"
+        )
+
+
 def test_encode_json():
     encoded_data = decoder_encoder.encode_json(ENCODE_JSON_INPUT)
     assert json.loads(encoded_data) == ENCODE_JSON_INPUT

tests/unit/test_handler_service.py

@@ -17,13 +17,14 @@
 import tempfile
 
 import pytest
+from sagemaker_inference import content_types
 from transformers.testing_utils import require_torch, slow
 
 from mms.context import Context, RequestProcessor
 from mms.metrics.metrics_store import MetricsStore
 from sagemaker_huggingface_inference_toolkit import handler_service
 from sagemaker_huggingface_inference_toolkit.transformers_utils import _load_model_from_hub, get_pipeline
-from sagemaker_inference import content_types
+
 
 TASK = "text-classification"
 MODEL = "sshleifer/tiny-dbmdz-bert-large-cased-finetuned-conll03-english"