Reformat python code

Author: siddvenk

src/sagemaker/djl_inference/defaults.py

@@ -42,4 +42,4 @@
     "ml.g5",
     "ml.p3",
     "ml.p4",
-}
+}

src/sagemaker/djl_inference/model.py

@@ -18,7 +18,7 @@
 from enum import Enum
 from typing import Optional, Union, Dict
 
-import defaults
+from sagemaker.djl_inference import defaults
 import sagemaker
 from sagemaker import s3, Predictor, image_uris, fw_utils
 from sagemaker.deserializers import JSONDeserializer
@@ -94,9 +94,11 @@ def __init__(
         **kwargs,
     ):
         if kwargs.get("model_data") is not None:
-            raise ValueError("DJLLargeModels do not support the model_data parameter. Please use"
-                             "uncompressed_model_data and ensure the s3 uri points to a folder containing"
-                             "all model artifacts, not a tar.gz file")
+            raise ValueError(
+                "DJLLargeModels do not support the model_data parameter. Please use"
+                "uncompressed_model_data and ensure the s3 uri points to a folder containing"
+                "all model artifacts, not a tar.gz file"
+            )
         super(DJLLargeModel, self).__init__(
             None, image_uri, role, entry_point, predictor_cls=predictor_cls, **kwargs
         )
@@ -126,6 +128,7 @@ def package_for_edge(
         tags=None,
     ):
         raise NotImplementedError("DJLLargeModels do not support Sagemaker Edge")
+
     def compile(
         self,
         target_instance_family,
@@ -142,7 +145,9 @@ def compile(
         target_platform_accelerator=None,
         compiler_options=None,
     ):
-        raise NotImplementedError("DJLLargeModels do not currently support compilation with SageMaker Neo")
+        raise NotImplementedError(
+            "DJLLargeModels do not currently support compilation with SageMaker Neo"
+        )
 
     def deploy(
         self,
@@ -168,14 +173,18 @@ def deploy(
         if serverless_inference_config:
             raise ValueError("DJLLargeModels do not support Serverless Deployment")
         if instance_type is None and not self.inference_recommender_job_results:
-            raise ValueError(f"instance_type must be specified, or inference recommendation from right_size()"
-                             "must be run to deploy the model. Supported instance type families are :"
-                             f"{defaults.ALLOWED_INSTANCE_FAMILIES}")
+            raise ValueError(
+                f"instance_type must be specified, or inference recommendation from right_size()"
+                "must be run to deploy the model. Supported instance type families are :"
+                f"{defaults.ALLOWED_INSTANCE_FAMILIES}"
+            )
         if instance_type:
-            instance_family = instance_type.rsplit('.', 1)[0]
+            instance_family = instance_type.rsplit(".", 1)[0]
             if not instance_family in defaults.ALLOWED_INSTANCE_FAMILIES:
-                raise ValueError(f"Invalid instance type. DJLLargeModels only support deployment to instances"
-                                 f"with GPUs. Supported instance families are {defaults.ALLOWED_INSTANCE_FAMILIES}")
+                raise ValueError(
+                    f"Invalid instance type. DJLLargeModels only support deployment to instances"
+                    f"with GPUs. Supported instance families are {defaults.ALLOWED_INSTANCE_FAMILIES}"
+                )
 
         super(DJLLargeModel, self).deploy(
             initial_instance_count=initial_instance_count,
@@ -253,8 +262,8 @@ def prepare_container_def(
             )
 
     def generate_serving_properties(self, serving_properties={}) -> Dict[str, str]:
-        serving_properties["engine"] = self.engine[0]
-        serving_properties["option.entryPoint"] = self.engine[1]
+        serving_properties["engine"] = self.engine.value[0]
+        serving_properties["option.entryPoint"] = self.engine.value[1]
         serving_properties["option.s3url"] = self.uncompressed_model_data
         if self.tensor_parallel_degree:
             serving_properties["option.tensor_parallel_degree"] = self.tensor_parallel_degree