feature: support create Clarify explainer enabled endpoint for Clarify Online Explainability (#3727)

Co-authored-by: Clayton Parnell <[email protected]>

Author: 2 people

doc/api/inference/explainer.rst

@@ -0,0 +1,16 @@
+Online Explainability
+---------------------
+
+This module contains classes related to Amazon Sagemaker Clarify Online Explainability
+
+.. automodule:: sagemaker.explainer.explainer_config
+    :members:
+    :undoc-members:
+    :show-inheritance:
+
+.. automodule:: sagemaker.explainer.clarify_explainer_config
+    :members:
+    :undoc-members:
+    :show-inheritance:
+
+

src/sagemaker/estimator.py

@@ -1378,6 +1378,7 @@ def deploy(
         model_data_download_timeout=None,
         container_startup_health_check_timeout=None,
         inference_recommendation_id=None,
+        explainer_config=None,
         **kwargs,
     ):
         """Deploy the trained model to an Amazon SageMaker endpoint.
@@ -1458,6 +1459,8 @@ def deploy(
             inference_recommendation_id (str): The recommendation id which specifies the
                 recommendation you picked from inference recommendation job results and
                 would like to deploy the model and endpoint with recommended parameters.
+            explainer_config (sagemaker.explainer.ExplainerConfig): Specifies online explainability
+                configuration for use with Amazon SageMaker Clarify. (default: None)
             **kwargs: Passed to invocation of ``create_model()``.
                 Implementations may customize ``create_model()`` to accept
                 ``**kwargs`` to customize model creation during deploy.
@@ -1516,6 +1519,7 @@ def deploy(
             data_capture_config=data_capture_config,
             serverless_inference_config=serverless_inference_config,
             async_inference_config=async_inference_config,
+            explainer_config=explainer_config,
             volume_size=volume_size,
             model_data_download_timeout=model_data_download_timeout,
             container_startup_health_check_timeout=container_startup_health_check_timeout,

src/sagemaker/explainer/__init__.py

@@ -0,0 +1,24 @@
+# Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://aws.amazon.com/apache2.0/
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, 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.
+"""Imports the classes in this module to simplify customer imports"""
+
+from __future__ import absolute_import
+
+from sagemaker.explainer.explainer_config import ExplainerConfig  # noqa: F401
+from sagemaker.explainer.clarify_explainer_config import (  # noqa: F401
+    ClarifyExplainerConfig,
+    ClarifyInferenceConfig,
+    ClarifyShapConfig,
+    ClarifyShapBaselineConfig,
+    ClarifyTextConfig,
+)

src/sagemaker/explainer/clarify_explainer_config.py

@@ -0,0 +1,44 @@
+# Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://aws.amazon.com/apache2.0/
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, 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.
+"""A member of ``CreateEndpointConfig`` that enables explainers."""
+
+from __future__ import print_function, absolute_import
+from typing import Optional
+from sagemaker.explainer.clarify_explainer_config import ClarifyExplainerConfig
+
+
+class ExplainerConfig(object):
+    """A parameter to activate explainers."""
+
+    def __init__(
+        self,
+        clarify_explainer_config: Optional[ClarifyExplainerConfig] = None,
+    ):
+        """Initializes a config object to activate explainer.
+
+        Args:
+            clarify_explainer_config (:class:`~sagemaker.explainer.explainer_config.ClarifyExplainerConfig`):
+                Optional. A config contains parameters for the SageMaker Clarify explainer. (Default: None)
+        """  # noqa E501  # pylint: disable=line-too-long
+        self.clarify_explainer_config = clarify_explainer_config
+
+    def _to_request_dict(self):
+        """Generates a request dictionary using the parameters provided to the class."""
+        request_dict = {}
+
+        if self.clarify_explainer_config:
+            request_dict[
+                "ClarifyExplainerConfig"
+            ] = self.clarify_explainer_config._to_request_dict()
+
+        return request_dict

src/sagemaker/explainer/explainer_config.py

@@ -210,6 +210,7 @@ def deploy(
         model_data_download_timeout=None,
         container_startup_health_check_timeout=None,
         inference_recommendation_id=None,
+        explainer_config=None,
         **kwargs,
     ):
         """Deploy this ``Model`` to an ``Endpoint`` and optionally return a ``Predictor``.
@@ -286,6 +287,8 @@ def deploy(
             inference_recommendation_id (str): The recommendation id which specifies the
                 recommendation you picked from inference recommendation job results and
                 would like to deploy the model and endpoint with recommended parameters.
+            explainer_config (sagemaker.explainer.ExplainerConfig): Specifies online explainability
+                configuration for use with Amazon SageMaker Clarify. (default: None)
         Raises:
              ValueError: If arguments combination check failed in these circumstances:
                 - If no role is specified or
@@ -322,6 +325,7 @@ def deploy(
             model_data_download_timeout=model_data_download_timeout,
             container_startup_health_check_timeout=container_startup_health_check_timeout,
             inference_recommendation_id=inference_recommendation_id,
+            explainer_config=explainer_config,
         )
 
     def register(

src/sagemaker/huggingface/model.py

@@ -215,6 +215,7 @@ def _update_params(
         accelerator_type = kwargs["accelerator_type"]
         async_inference_config = kwargs["async_inference_config"]
         serverless_inference_config = kwargs["serverless_inference_config"]
+        explainer_config = kwargs["explainer_config"]
         inference_recommendation_id = kwargs["inference_recommendation_id"]
         inference_recommender_job_results = kwargs["inference_recommender_job_results"]
         if inference_recommendation_id is not None:
@@ -225,6 +226,7 @@ def _update_params(
                 async_inference_config=async_inference_config,
                 serverless_inference_config=serverless_inference_config,
                 inference_recommendation_id=inference_recommendation_id,
+                explainer_config=explainer_config,
             )
         elif inference_recommender_job_results is not None:
             inference_recommendation = self._update_params_for_right_size(
@@ -233,6 +235,7 @@ def _update_params(
                 accelerator_type,
                 serverless_inference_config,
                 async_inference_config,
+                explainer_config,
             )
         return inference_recommendation or (instance_type, initial_instance_count)
 
@@ -243,6 +246,7 @@ def _update_params_for_right_size(
         accelerator_type=None,
         serverless_inference_config=None,
         async_inference_config=None,
+        explainer_config=None,
     ):
         """Validates that Inference Recommendation parameters can be used in `model.deploy()`
 
@@ -262,6 +266,8 @@ def _update_params_for_right_size(
                 whether serverless_inference_config has been passed into `model.deploy()`.
             async_inference_config (sagemaker.model_monitor.AsyncInferenceConfig):
                 whether async_inference_config has been passed into `model.deploy()`.
+            explainer_config (sagemaker.explainer.ExplainerConfig): whether explainer_config
+                has been passed into `model.deploy()`.
 
         Returns:
             (string, int) or None: Top instance_type and associated initial_instance_count
@@ -285,6 +291,11 @@ def _update_params_for_right_size(
                 "serverless_inference_config is specified. Overriding right_size() recommendations."
             )
             return None
+        if explainer_config:
+            LOGGER.warning(
+                "explainer_config is specified. Overriding right_size() recommendations."
+            )
+            return None
 
         instance_type = self.inference_recommendations[0]["EndpointConfiguration"]["InstanceType"]
         initial_instance_count = self.inference_recommendations[0]["EndpointConfiguration"][
@@ -300,6 +311,7 @@ def _update_params_for_recommendation_id(
         async_inference_config,
         serverless_inference_config,
         inference_recommendation_id,
+        explainer_config,
     ):
         """Update parameters with inference recommendation results.
 
@@ -332,6 +344,8 @@ def _update_params_for_recommendation_id(
                 the recommendation you picked from inference recommendation job
                 results and would like to deploy the model and endpoint with
                 recommended parameters.
+            explainer_config (sagemaker.explainer.ExplainerConfig): Specifies online explainability
+                configuration for use with Amazon SageMaker Clarify. Default: None.
         Raises:
             ValueError: If arguments combination check failed in these circumstances:
                 - If only one of instance type or instance count specified or
@@ -367,6 +381,8 @@ def _update_params_for_recommendation_id(
             raise ValueError(
                 "serverless_inference_config is not compatible with inference_recommendation_id."
             )
+        if explainer_config is not None:
+            raise ValueError("explainer_config is not compatible with inference_recommendation_id.")
 
         # Validate recommendation id
         if not re.match(r"[a-zA-Z0-9](-*[a-zA-Z0-9]){0,63}\/\w{8}$", inference_recommendation_id):

src/sagemaker/inference_recommender/inference_recommender_mixin.py

@@ -42,6 +42,7 @@
 from sagemaker.model_metrics import ModelMetrics
 from sagemaker.deprecations import removed_kwargs
 from sagemaker.drift_check_baselines import DriftCheckBaselines
+from sagemaker.explainer import ExplainerConfig
 from sagemaker.metadata_properties import MetadataProperties
 from sagemaker.predictor import PredictorBase
 from sagemaker.serverless import ServerlessInferenceConfig
@@ -1080,6 +1081,7 @@ def deploy(
         model_data_download_timeout=None,
         container_startup_health_check_timeout=None,
         inference_recommendation_id=None,
+        explainer_config=None,
         **kwargs,
     ):
         """Deploy this ``Model`` to an ``Endpoint`` and optionally return a ``Predictor``.
@@ -1158,6 +1160,8 @@ def deploy(
             inference_recommendation_id (str): The recommendation id which specifies the
                 recommendation you picked from inference recommendation job results and
                 would like to deploy the model and endpoint with recommended parameters.
+            explainer_config (sagemaker.explainer.ExplainerConfig): Specifies online explainability
+                configuration for use with Amazon SageMaker Clarify. Default: None.
         Raises:
              ValueError: If arguments combination check failed in these circumstances:
                 - If no role is specified or
@@ -1204,6 +1208,7 @@ def deploy(
                 accelerator_type=accelerator_type,
                 async_inference_config=async_inference_config,
                 serverless_inference_config=serverless_inference_config,
+                explainer_config=explainer_config,
                 inference_recommendation_id=inference_recommendation_id,
                 inference_recommender_job_results=self.inference_recommender_job_results,
             )
@@ -1212,6 +1217,10 @@ def deploy(
         if is_async and not isinstance(async_inference_config, AsyncInferenceConfig):
             raise ValueError("async_inference_config needs to be a AsyncInferenceConfig object")
 
+        is_explainer_enabled = explainer_config is not None
+        if is_explainer_enabled and not isinstance(explainer_config, ExplainerConfig):
+            raise ValueError("explainer_config needs to be a ExplainerConfig object")
+
         is_serverless = serverless_inference_config is not None
         if not is_serverless and not (instance_type and initial_instance_count):
             raise ValueError(
@@ -1282,13 +1291,18 @@ def deploy(
             )
             async_inference_config_dict = async_inference_config._to_request_dict()
 
+        explainer_config_dict = None
+        if is_explainer_enabled:
+            explainer_config_dict = explainer_config._to_request_dict()
+
         self.sagemaker_session.endpoint_from_production_variants(
             name=self.endpoint_name,
             production_variants=[production_variant],
             tags=tags,
             kms_key=kms_key,
             wait=wait,
             data_capture_config_dict=data_capture_config_dict,
+            explainer_config_dict=explainer_config_dict,
             async_inference_config_dict=async_inference_config_dict,
         )
 

src/sagemaker/model.py

@@ -3663,6 +3663,7 @@ def create_endpoint_config(
         volume_size=None,
         model_data_download_timeout=None,
         container_startup_health_check_timeout=None,
+        explainer_config_dict=None,
     ):
         """Create an Amazon SageMaker endpoint configuration.
 
@@ -3696,6 +3697,8 @@ def create_endpoint_config(
                 inference container to pass health check by SageMaker Hosting. For more information
                 about health check see:
                 https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms-inference-code.html#your-algorithms-inference-algo-ping-requests
+            explainer_config_dict (dict): Specifies configuration to enable explainers.
+                Default: None.
 
         Example:
             >>> tags = [{'Key': 'tagname', 'Value': 'tagvalue'}]
@@ -3751,6 +3754,9 @@ def create_endpoint_config(
             )
             request["DataCaptureConfig"] = inferred_data_capture_config_dict
 
+        if explainer_config_dict is not None:
+            request["ExplainerConfig"] = explainer_config_dict
+
         self.sagemaker_client.create_endpoint_config(**request)
         return name
 
@@ -3762,6 +3768,7 @@ def create_endpoint_config_from_existing(
         new_kms_key=None,
         new_data_capture_config_dict=None,
         new_production_variants=None,
+        new_explainer_config_dict=None,
     ):
         """Create an Amazon SageMaker endpoint configuration from an existing one.
 
@@ -3789,6 +3796,9 @@ def create_endpoint_config_from_existing(
             new_production_variants (list[dict]): The configuration for which model(s) to host and
                 the resources to deploy for hosting the model(s). If not specified,
                 the ``ProductionVariants`` of the existing endpoint configuration is used.
+            new_explainer_config_dict (dict): Specifies configuration to enable explainers.
+                (default: None). If not specified, the explainer configuration of the existing
+                endpoint configuration is used.
 
         Returns:
             str: Name of the endpoint point configuration created.
@@ -3856,6 +3866,13 @@ def create_endpoint_config_from_existing(
             )
             request["AsyncInferenceConfig"] = inferred_async_inference_config_dict
 
+        request_explainer_config_dict = (
+            new_explainer_config_dict or existing_endpoint_config_desc.get("ExplainerConfig", None)
+        )
+
+        if request_explainer_config_dict is not None:
+            request["ExplainerConfig"] = request_explainer_config_dict
+
         self.sagemaker_client.create_endpoint_config(**request)
 
     def create_endpoint(self, endpoint_name, config_name, tags=None, wait=True):
@@ -4372,6 +4389,7 @@ def endpoint_from_production_variants(
         wait=True,
         data_capture_config_dict=None,
         async_inference_config_dict=None,
+        explainer_config_dict=None,
     ):
         """Create an SageMaker ``Endpoint`` from a list of production variants.
 
@@ -4389,6 +4407,9 @@ def endpoint_from_production_variants(
             async_inference_config_dict (dict) : specifies configuration related to async endpoint.
                 Use this configuration when trying to create async endpoint and make async inference
                 (default: None)
+            explainer_config_dict (dict) : Specifies configuration related to explainer.
+                Use this configuration when trying to use online explainability.
+                (default: None)
         Returns:
             str: The name of the created ``Endpoint``.
         """
@@ -4422,6 +4443,8 @@ def endpoint_from_production_variants(
                 sagemaker_session=self,
             )
             config_options["AsyncInferenceConfig"] = inferred_async_inference_config_dict
+        if explainer_config_dict is not None:
+            config_options["ExplainerConfig"] = explainer_config_dict
 
         LOGGER.info("Creating endpoint-config with name %s", name)
         self.sagemaker_client.create_endpoint_config(**config_options)

src/sagemaker/session.py

@@ -324,6 +324,7 @@ def deploy(
         model_data_download_timeout=None,
         container_startup_health_check_timeout=None,
         inference_recommendation_id=None,
+        explainer_config=None,
     ):
         """Deploy a Tensorflow ``Model`` to a SageMaker ``Endpoint``."""
 
@@ -349,6 +350,7 @@ def deploy(
             container_startup_health_check_timeout=container_startup_health_check_timeout,
             update_endpoint=update_endpoint,
             inference_recommendation_id=inference_recommendation_id,
+            explainer_config=explainer_config,
         )
 
     def _eia_supported(self):