Merge branch 'master' into tf242ioc

Author: navinsoni

CHANGELOG.md

@@ -1,5 +1,24 @@
 # Changelog
 
+## v2.81.0 (2022-03-26)
+
+### Features
+
+ * Retrieve data configuration
+ * enable EnableInterContainerTrafficEncryption for model monitoring
+ * Hugging Face Transformers 4.17 for PT 1.10
+
+### Bug Fixes and Other Changes
+
+ * remove `new` from serverless
+ * temporarily skip tests impacted by data inconsistency
+ * Implement override solution for pipeline variables
+
+### Documentation Changes
+
+ * add documentation for image_uri serverless use case
+ * minor fixes for smddp 1.4.0 doc
+
 ## v2.80.0 (2022-03-18)
 
 ### Features

VERSION

@@ -1 +1 @@
-2.80.1.dev0
+2.81.1.dev0

doc/overview.rst

@@ -1226,6 +1226,28 @@ to configure or manage the underlying infrastructure. After you trained a model,
 Serverless endpoint and then invoke the endpoint with the model to get inference results back. More information about
 SageMaker Serverless Inference can be found in the `AWS documentation <https://docs.aws.amazon.com/sagemaker/latest/dg/serverless-endpoints.html>`__.
 
+For using SageMaker Serverless Inference, if you plan to use any of the SageMaker-provided container or Bring Your Own Container
+model, you will need to pass ``image_uri``. An example to use ``image_uri`` for creating MXNet model:
+
+.. code:: python
+
+    from sagemaker.mxnet import MXNetModel
+    import sagemaker
+
+    role = sagemaker.get_execution_role()
+
+    # create MXNet Model Class
+    mxnet_model = MXNetModel(
+        model_data="s3://my_bucket/pretrained_model/model.tar.gz", # path to your trained sagemaker model
+        role=role, # iam role with permissions to create an Endpoint
+        entry_point="inference.py",
+        image_uri="763104351884.dkr.ecr.us-west-2.amazonaws.com/mxnet-inference:1.4.1-cpu-py3" # image wanted to use
+    )
+
+For more Amazon SageMaker provided algorithms and containers image paths, please check this page: `Amazon SageMaker provided
+algorithms and Deep Learning Containers <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+After creating model using ``image_uri``, you can then follow the steps below to create serverless endpoint.
+
 To deploy serverless endpoint, you will need to create a ``ServerlessInferenceConfig``.
 If you create ``ServerlessInferenceConfig`` without specifying its arguments, the default ``MemorySizeInMB`` will be **2048** and
 the default ``MaxConcurrency`` will be **5** :
@@ -1235,14 +1257,14 @@ the default ``MaxConcurrency`` will be **5** :
     from sagemaker.serverless import ServerlessInferenceConfig
 
     # Create an empty ServerlessInferenceConfig object to use default values
-    serverless_config = new ServerlessInferenceConfig()
+    serverless_config = ServerlessInferenceConfig()
 
 Or you can specify ``MemorySizeInMB`` and ``MaxConcurrency`` in ``ServerlessInferenceConfig`` (example shown below):
 
 .. code:: python
 
     # Specify MemorySizeInMB and MaxConcurrency in the serverless config object
-    serverless_config = new ServerlessInferenceConfig(
+    serverless_config = ServerlessInferenceConfig(
       memory_size_in_mb=4096,
       max_concurrency=10,
     )
@@ -1254,6 +1276,14 @@ Then use the ``ServerlessInferenceConfig`` in the estimator's ``deploy()`` metho
     # Deploys the model that was generated by fit() to a SageMaker serverless endpoint
     serverless_predictor = estimator.deploy(serverless_inference_config=serverless_config)
 
+Or directly using model's ``deploy()`` method to deploy a serverless endpoint:
+
+.. code:: python
+
+    # Deploys the model to a SageMaker serverless endpoint
+    serverless_predictor = model.deploy(serverless_inference_config=serverless_config)
+
+
 After deployment is complete, you can use predictor's ``predict()`` method to invoke the serverless endpoint just like
 real-time endpoints:
 

setup.py

@@ -16,7 +16,7 @@
 import os
 from glob import glob
 
-from setuptools import setup, find_packages
+from setuptools import find_packages, setup
 
 
 def read(fname):
@@ -81,6 +81,7 @@ def read_version():
         "fabric==2.6.0",
         "requests==2.27.1",
         "sagemaker-experiments==0.1.35",
+        "Jinja2==3.0.3",
     ],
 )
 

src/sagemaker/chainer/model.py

@@ -99,10 +99,14 @@ def __init__(
                 file which should be executed as the entry point to model
                 hosting. If ``source_dir`` is specified, then ``entry_point``
                 must point to a file located at the root of ``source_dir``.
-            image_uri (str): A Docker image URI (default: None). If not specified, a
-                default image for Chainer will be used. If ``framework_version``
-                or ``py_version`` are ``None``, then ``image_uri`` is required. If
-                also ``None``, then a ``ValueError`` will be raised.
+            image_uri (str): A Docker image URI (default: None). In serverless
+                inferece, it is required. More image information can be found in
+                `Amazon SageMaker provided algorithms and Deep Learning Containers
+                <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+                In instance based inference, if not specified, a default image for
+                Chainer will be used. If ``framework_version`` or ``py_version``
+                are ``None``, then ``image_uri`` is required. If also ``None``,
+                then a ``ValueError`` will be raised.
             framework_version (str): Chainer version you want to use for
                 executing your model training code. Defaults to ``None``. Required
                 unless ``image_uri`` is provided.

src/sagemaker/estimator.py

@@ -74,9 +74,7 @@
     get_config_value,
     name_from_base,
 )
-from sagemaker.workflow.entities import Expression
-from sagemaker.workflow.parameters import Parameter
-from sagemaker.workflow.properties import Properties
+from sagemaker.workflow.entities import PipelineVariable
 
 logger = logging.getLogger(__name__)
 
@@ -602,7 +600,7 @@ def _json_encode_hyperparameters(hyperparameters: Dict[str, Any]) -> Dict[str, A
         current_hyperparameters = hyperparameters
         if current_hyperparameters is not None:
             hyperparameters = {
-                str(k): (v if isinstance(v, (Parameter, Expression, Properties)) else json.dumps(v))
+                str(k): (v.to_string() if isinstance(v, PipelineVariable) else json.dumps(v))
                 for (k, v) in current_hyperparameters.items()
             }
         return hyperparameters
@@ -1813,7 +1811,7 @@ def _get_train_args(cls, estimator, inputs, experiment_config):
         current_hyperparameters = estimator.hyperparameters()
         if current_hyperparameters is not None:
             hyperparameters = {
-                str(k): (v if isinstance(v, (Parameter, Expression, Properties)) else str(v))
+                str(k): (v.to_string() if isinstance(v, PipelineVariable) else str(v))
                 for (k, v) in current_hyperparameters.items()
             }
 

src/sagemaker/huggingface/model.py

@@ -133,7 +133,11 @@ def __init__(
             py_version (str): Python version you want to use for executing your
                 model training code. Defaults to ``None``. Required unless
                 ``image_uri`` is provided.
-            image_uri (str): A Docker image URI. Defaults to None. If not specified, a
+            image_uri (str): A Docker image URI. Defaults to None. For serverless
+                inferece, it is required. More image information can be found in
+                `Amazon SageMaker provided algorithms and Deep Learning Containers
+                <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+                For instance based inference, if not specified, a
                 default image for PyTorch will be used. If ``framework_version``
                 or ``py_version`` are ``None``, then ``image_uri`` is required. If
                 also ``None``, then a ``ValueError`` will be raised.

src/sagemaker/model_monitor/clarify_model_monitoring.py

@@ -397,7 +397,6 @@ def _build_create_job_definition_request(
 
         if network_config is not None:
             network_config_dict = network_config._to_request_dict()
-            self._validate_network_config(network_config_dict)
             request_dict["NetworkConfig"] = network_config_dict
         elif existing_network_config is not None:
             request_dict["NetworkConfig"] = existing_network_config

src/sagemaker/model_monitor/model_monitoring.py

@@ -295,7 +295,6 @@ def create_monitoring_schedule(
         network_config_dict = None
         if self.network_config is not None:
             network_config_dict = self.network_config._to_request_dict()
-            self._validate_network_config(network_config_dict)
 
         self.sagemaker_session.create_monitoring_schedule(
             monitoring_schedule_name=self.monitoring_schedule_name,
@@ -448,7 +447,6 @@ def update_monitoring_schedule(
         network_config_dict = None
         if self.network_config is not None:
             network_config_dict = self.network_config._to_request_dict()
-            self._validate_network_config(network_config_dict)
 
         self.sagemaker_session.update_monitoring_schedule(
             monitoring_schedule_name=self.monitoring_schedule_name,
@@ -708,6 +706,9 @@ def attach(cls, monitor_schedule_name, sagemaker_session=None):
         if network_config_dict:
             network_config = NetworkConfig(
                 enable_network_isolation=network_config_dict["EnableNetworkIsolation"],
+                encrypt_inter_container_traffic=network_config_dict[
+                    "EnableInterContainerTrafficEncryption"
+                ],
                 security_group_ids=security_group_ids,
                 subnets=subnets,
             )
@@ -784,6 +785,9 @@ def _attach(clazz, sagemaker_session, schedule_desc, job_desc, tags):
         if network_config_dict:
             network_config = NetworkConfig(
                 enable_network_isolation=network_config_dict["EnableNetworkIsolation"],
+                encrypt_inter_container_traffic=network_config_dict[
+                    "EnableInterContainerTrafficEncryption"
+                ],
                 security_group_ids=security_group_ids,
                 subnets=subnets,
             )
@@ -1164,31 +1168,6 @@ def _wait_for_schedule_changes_to_apply(self):
             if schedule_desc["MonitoringScheduleStatus"] != "Pending":
                 break
 
-    def _validate_network_config(self, network_config_dict):
-        """Function to validate EnableInterContainerTrafficEncryption.
-
-        It validates EnableInterContainerTrafficEncryption is not set in the provided
-        NetworkConfig request dictionary.
-
-        Args:
-            network_config_dict (dict): NetworkConfig request dictionary.
-                Contains parameters from :class:`~sagemaker.network.NetworkConfig` object
-                that configures network isolation, encryption of
-                inter-container traffic, security group IDs, and subnets.
-
-        """
-        if "EnableInterContainerTrafficEncryption" in network_config_dict:
-            message = (
-                "EnableInterContainerTrafficEncryption is not supported in Model Monitor. "
-                "Please ensure that encrypt_inter_container_traffic=None "
-                "when creating your NetworkConfig object. "
-                "Current encrypt_inter_container_traffic value: {}".format(
-                    self.network_config.encrypt_inter_container_traffic
-                )
-            )
-            _LOGGER.info(message)
-            raise ValueError(message)
-
     @classmethod
     def monitoring_type(cls):
         """Type of the monitoring job."""
@@ -1781,7 +1760,6 @@ def update_monitoring_schedule(
         network_config_dict = None
         if self.network_config is not None:
             network_config_dict = self.network_config._to_request_dict()
-            super(DefaultModelMonitor, self)._validate_network_config(network_config_dict)
 
         if role is not None:
             self.role = role
@@ -2034,6 +2012,9 @@ def attach(cls, monitor_schedule_name, sagemaker_session=None):
             subnets = vpc_config.get("Subnets")
             network_config = NetworkConfig(
                 enable_network_isolation=network_config_dict["EnableNetworkIsolation"],
+                encrypt_inter_container_traffic=network_config_dict[
+                    "EnableInterContainerTrafficEncryption"
+                ],
                 security_group_ids=security_group_ids,
                 subnets=subnets,
             )
@@ -2304,7 +2285,6 @@ def _build_create_data_quality_job_definition_request(
 
         if network_config is not None:
             network_config_dict = network_config._to_request_dict()
-            self._validate_network_config(network_config_dict)
             request_dict["NetworkConfig"] = network_config_dict
         elif existing_network_config is not None:
             request_dict["NetworkConfig"] = existing_network_config
@@ -3007,7 +2987,6 @@ def _build_create_model_quality_job_definition_request(
 
         if network_config is not None:
             network_config_dict = network_config._to_request_dict()
-            self._validate_network_config(network_config_dict)
             request_dict["NetworkConfig"] = network_config_dict
         elif existing_network_config is not None:
             request_dict["NetworkConfig"] = existing_network_config

src/sagemaker/mxnet/model.py

@@ -107,9 +107,12 @@ def __init__(
             py_version (str): Python version you want to use for executing your
                 model training code. Defaults to ``None``. Required unless
                 ``image_uri`` is provided.
-            image_uri (str): A Docker image URI (default: None). If not specified, a
-                default image for MXNet will be used.
-
+            image_uri (str): A Docker image URI (default: None). For serverless
+                inferece, it is required. More image information can be found in
+                `Amazon SageMaker provided algorithms and Deep Learning Containers
+                <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+                For instance based inference, if not specified, a default image for
+                MXNet will be used.
                 If ``framework_version`` or ``py_version`` are ``None``, then
                 ``image_uri`` is required. If also ``None``, then a ``ValueError``
                 will be raised.

src/sagemaker/parameter.py

@@ -14,9 +14,8 @@
 from __future__ import absolute_import
 
 import json
-from sagemaker.workflow.parameters import Parameter as PipelineParameter
-from sagemaker.workflow.functions import JsonGet as PipelineJsonGet
-from sagemaker.workflow.functions import Join as PipelineJoin
+
+from sagemaker.workflow.entities import PipelineVariable
 
 
 class ParameterRange(object):
@@ -73,11 +72,11 @@ def as_tuning_range(self, name):
         return {
             "Name": name,
             "MinValue": str(self.min_value)
-            if not isinstance(self.min_value, (PipelineParameter, PipelineJsonGet, PipelineJoin))
-            else self.min_value,
+            if not isinstance(self.min_value, PipelineVariable)
+            else self.min_value.to_string(),
             "MaxValue": str(self.max_value)
-            if not isinstance(self.max_value, (PipelineParameter, PipelineJsonGet, PipelineJoin))
-            else self.max_value,
+            if not isinstance(self.max_value, PipelineVariable)
+            else self.max_value.to_string(),
             "ScalingType": self.scaling_type,
         }
 
@@ -112,8 +111,7 @@ def __init__(self, values):  # pylint: disable=super-init-not-called
         """
         values = values if isinstance(values, list) else [values]
         self.values = [
-            str(v) if not isinstance(v, (PipelineParameter, PipelineJsonGet, PipelineJoin)) else v
-            for v in values
+            str(v) if not isinstance(v, PipelineVariable) else v.to_string() for v in values
         ]
 
     def as_tuning_range(self, name):

src/sagemaker/pytorch/model.py

@@ -107,10 +107,14 @@ def __init__(
             py_version (str): Python version you want to use for executing your
                 model training code. Defaults to ``None``. Required unless
                 ``image_uri`` is provided.
-            image_uri (str): A Docker image URI (default: None). If not specified, a
-                default image for PyTorch will be used. If ``framework_version``
-                or ``py_version`` are ``None``, then ``image_uri`` is required. If
-                also ``None``, then a ``ValueError`` will be raised.
+            image_uri (str): A Docker image URI (default: None). For serverless
+                inferece, it is required. More image information can be found in
+                `Amazon SageMaker provided algorithms and Deep Learning Containers
+                <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+                For instance based inference, if not specified, a default image for
+                PyTorch will be used. If ``framework_version`` or ``py_version`` are
+                ``None``, then ``image_uri`` is required. If also ``None``, then a
+                ``ValueError`` will be raised.
             predictor_cls (callable[str, sagemaker.session.Session]): A function
                 to call to create a predictor with an endpoint name and
                 SageMaker ``Session``. If specified, ``deploy()`` returns the

src/sagemaker/sklearn/model.py

@@ -102,9 +102,12 @@ def __init__(
                 model training code (default: 'py3'). Currently, 'py3' is the only
                 supported version. If ``None`` is passed in, ``image_uri`` must be
                 provided.
-            image_uri (str): A Docker image URI (default: None). If not specified, a
-                default image for Scikit-learn will be used.
-
+            image_uri (str): A Docker image URI (default: None). For serverless
+                inferece, it is required. More image information can be found in
+                `Amazon SageMaker provided algorithms and Deep Learning Containers
+                <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+                For instance based inference, if not specified, a default image for
+                Scikit-learn will be used.
                 If ``framework_version`` or ``py_version`` are ``None``, then
                 ``image_uri`` is required. If also ``None``, then a ``ValueError``
                 will be raised.

src/sagemaker/tensorflow/model.py

@@ -145,10 +145,14 @@ def __init__(
                 file which should be executed as the entry point to model
                 hosting. If ``source_dir`` is specified, then ``entry_point``
                 must point to a file located at the root of ``source_dir``.
-            image_uri (str): A Docker image URI (default: None). If not specified, a
-                default image for TensorFlow Serving will be used. If
-                ``framework_version`` is ``None``, then ``image_uri`` is required.
-                If also ``None``, then a ``ValueError`` will be raised.
+            image_uri (str): A Docker image URI (default: None). For serverless
+                inferece, it is required. More image information can be found in
+                `Amazon SageMaker provided algorithms and Deep Learning Containers
+                <https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-algo-docker-registry-paths.html>`_.
+                For instance based inference, if not specified, a default image for
+                TensorFlow Serving will be used. If ``framework_version`` is ``None``,
+                then ``image_uri`` is required. If also ``None``, then a ``ValueError``
+                will be raised.
             framework_version (str): Optional. TensorFlow Serving version you
                 want to use. Defaults to ``None``. Required unless ``image_uri`` is
                 provided.