Add JSONDeserializer

Author: Balaji Veeramani

doc/frameworks/tensorflow/upgrade_from_legacy.rst

@@ -245,11 +245,11 @@ For example, if you want to use JSON serialization and deserialization:
 
 .. code:: python
 
-    from sagemaker.predictor import json_deserializer, json_serializer
+    from sagemaker.predictor import json_serializer
+    from sagemaker.deserializers import JSONDeserializer
 
     predictor.content_type = "application/json"
     predictor.serializer = json_serializer
-    predictor.accept = "application/json"
-    predictor.deserializer = json_deserializer
+    predictor.deserializer = JSONDeserializer()
 
     predictor.predict(data)

src/sagemaker/amazon/ipinsights.py

@@ -16,7 +16,8 @@
 from sagemaker.amazon.amazon_estimator import AmazonAlgorithmEstimatorBase, registry
 from sagemaker.amazon.hyperparameter import Hyperparameter as hp  # noqa
 from sagemaker.amazon.validation import ge, le
-from sagemaker.predictor import Predictor, csv_serializer, json_deserializer
+from sagemaker.deserializers import JSONDeserializer
+from sagemaker.predictor import Predictor, csv_serializer
 from sagemaker.model import Model
 from sagemaker.session import Session
 from sagemaker.vpc_utils import VPC_CONFIG_DEFAULT
@@ -198,7 +199,7 @@ def __init__(self, endpoint_name, sagemaker_session=None):
             endpoint_name,
             sagemaker_session,
             serializer=csv_serializer,
-            deserializer=json_deserializer,
+            deserializer=JSONDeserializer(),
         )
 
 

src/sagemaker/deserializers.py

@@ -154,3 +154,24 @@ def deserialize(self, data, content_type):
             data.close()
 
         raise ValueError("%s cannot read content type %s." % (__class__.__name__, content_type))
+
+
+class JSONDeserializer(BaseDeserializer):
+    """Deserialize data from an inference endpoint into a Python dictionary."""
+
+    ACCEPT = "application/json"
+
+    def deserialize(self, data, content_type):
+        """Deserialize data from an inference endpoint into a Python dictionary.
+
+        Args:
+            data (botocore.response.StreamingBody): Data to be deserialized.
+            content_type (str): The MIME type of the data.
+
+        Returns:
+            dict: The data deserialized into a Python dictionary.
+        """
+        try:
+            return json.load(codecs.getreader("utf-8")(data))
+        finally:
+            data.close()

src/sagemaker/mxnet/model.py

@@ -18,6 +18,7 @@
 import packaging.version
 
 import sagemaker
+from sagemaker.deserializers import JSONDeserializer
 from sagemaker.fw_utils import (
     create_image_uri,
     model_code_key_prefix,
@@ -26,7 +27,7 @@
 )
 from sagemaker.model import FrameworkModel, MODEL_SERVER_WORKERS_PARAM_NAME
 from sagemaker.mxnet import defaults
-from sagemaker.predictor import Predictor, json_serializer, json_deserializer
+from sagemaker.predictor import Predictor, json_serializer
 
 logger = logging.getLogger("sagemaker")
 
@@ -50,7 +51,7 @@ def __init__(self, endpoint_name, sagemaker_session=None):
                 using the default AWS configuration chain.
         """
         super(MXNetPredictor, self).__init__(
-            endpoint_name, sagemaker_session, json_serializer, json_deserializer
+            endpoint_name, sagemaker_session, json_serializer, JSONDeserializer()
         )
 
 

src/sagemaker/predictor.py

@@ -672,32 +672,6 @@ def _json_serialize_from_buffer(buff):
     return buff.read()
 
 
-class _JsonDeserializer(object):
-    """Placeholder docstring"""
-
-    def __init__(self):
-        """Placeholder docstring"""
-        self.accept = CONTENT_TYPE_JSON
-
-    def __call__(self, stream, content_type):
-        """Decode a JSON object into the corresponding Python object.
-
-        Args:
-            stream (stream): The response stream to be deserialized.
-            content_type (str): The content type of the response.
-
-        Returns:
-            object: Body of the response deserialized into a JSON object.
-        """
-        try:
-            return json.load(codecs.getreader("utf-8")(stream))
-        finally:
-            stream.close()
-
-
-json_deserializer = _JsonDeserializer()
-
-
 class _NPYSerializer(object):
     """Placeholder docstring"""
 

src/sagemaker/tensorflow/model.py

@@ -17,8 +17,9 @@
 
 import sagemaker
 from sagemaker.content_types import CONTENT_TYPE_JSON
+from sagemaker.deserializers import JSONDeserializer
 from sagemaker.fw_utils import create_image_uri
-from sagemaker.predictor import json_serializer, json_deserializer, Predictor
+from sagemaker.predictor import json_serializer, Predictor
 
 
 class TensorFlowPredictor(Predictor):
@@ -31,7 +32,7 @@ def __init__(
         endpoint_name,
         sagemaker_session=None,
         serializer=json_serializer,
-        deserializer=json_deserializer,
+        deserializer=JSONDeserializer(),
         content_type=None,
         model_name=None,
         model_version=None,

tests/integ/test_byo_estimator.py

@@ -86,7 +86,7 @@ def test_byo_estimator(sagemaker_session, region, cpu_instance_type, training_se
         predictor = model.deploy(1, cpu_instance_type, endpoint_name=job_name)
         predictor.serializer = fm_serializer
         predictor.content_type = "application/json"
-        predictor.deserializer = sagemaker.predictor.json_deserializer
+        predictor.deserializer = sagemaker.deserializers.JSONDeserializer()
 
         result = predictor.predict(training_set[0][:10])
 
@@ -132,7 +132,7 @@ def test_async_byo_estimator(sagemaker_session, region, cpu_instance_type, train
         predictor = model.deploy(1, cpu_instance_type, endpoint_name=endpoint_name)
         predictor.serializer = fm_serializer
         predictor.content_type = "application/json"
-        predictor.deserializer = sagemaker.predictor.json_deserializer
+        predictor.deserializer = sagemaker.deserializers.JSONDeserializer()
 
         result = predictor.predict(training_set[0][:10])
 

tests/integ/test_tfs.py

@@ -188,7 +188,7 @@ def test_predict_jsons_json_content_type(tfs_predictor):
         tfs_predictor.endpoint_name,
         tfs_predictor.sagemaker_session,
         serializer=None,
-        deserializer=sagemaker.predictor.json_deserializer,
+        deserializer=sagemaker.deserializers.JSONDeserializer(),
         content_type="application/json",
         accept="application/json",
     )
@@ -205,7 +205,7 @@ def test_predict_jsons(tfs_predictor):
         tfs_predictor.endpoint_name,
         tfs_predictor.sagemaker_session,
         serializer=None,
-        deserializer=sagemaker.predictor.json_deserializer,
+        deserializer=sagemaker.predictor.JSONDeserializer(),
         content_type="application/jsons",
         accept="application/jsons",
     )
@@ -222,7 +222,7 @@ def test_predict_jsonlines(tfs_predictor):
         tfs_predictor.endpoint_name,
         tfs_predictor.sagemaker_session,
         serializer=None,
-        deserializer=sagemaker.predictor.json_deserializer,
+        deserializer=sagemaker.predictor.JSONDeserializer(),
         content_type="application/jsonlines",
         accept="application/jsonlines",
     )

tests/integ/test_tuner.py

@@ -25,9 +25,9 @@
 from sagemaker.amazon.amazon_estimator import get_image_uri
 from sagemaker.amazon.common import read_records
 from sagemaker.chainer import Chainer
+from sagemaker.deserializers import JSONDeserializer
 from sagemaker.estimator import Estimator
 from sagemaker.mxnet.estimator import MXNet
-from sagemaker.predictor import json_deserializer
 from sagemaker.pytorch import PyTorch
 from sagemaker.tensorflow import TensorFlow
 from sagemaker.tuner import (
@@ -891,7 +891,7 @@ def test_tuning_byo_estimator(sagemaker_session, cpu_instance_type):
         predictor = tuner.deploy(1, cpu_instance_type, endpoint_name=best_training_job)
         predictor.serializer = _fm_serializer
         predictor.content_type = "application/json"
-        predictor.deserializer = json_deserializer
+        predictor.deserializer = JSONDeserializer()
 
         result = predictor.predict(datasets.one_p_mnist()[0][:10])
 

tests/integ/test_tuner_multi_algo.py

@@ -21,8 +21,8 @@
 from sagemaker.amazon.amazon_estimator import get_image_uri
 from sagemaker.analytics import HyperparameterTuningJobAnalytics
 from sagemaker.content_types import CONTENT_TYPE_JSON
+from sagemaker.deserializers import JSONDeserializer
 from sagemaker.estimator import Estimator
-from sagemaker.predictor import json_deserializer
 from sagemaker.tuner import ContinuousParameter, IntegerParameter, HyperparameterTuner
 from tests.integ import datasets, DATA_DIR, TUNING_DEFAULT_TIMEOUT_MINUTES
 from tests.integ.timeout import timeout, timeout_and_delete_endpoint_by_name
@@ -219,7 +219,7 @@ def _create_training_inputs(sagemaker_session):
 def _make_prediction(predictor, data):
     predictor.serializer = _prediction_data_serializer
     predictor.content_type = CONTENT_TYPE_JSON
-    predictor.deserializer = json_deserializer
+    predictor.deserializer = JSONDeserializer()
     return predictor.predict(data)