Merge remote-tracking branch 'origin/master'

# Conflicts:
#	tests/data/serve_resources/mlflow/pytorch/requirements.txt
#	tests/data/serve_resources/mlflow/xgboost/requirements.txt

Author: nargokul

.github/workflows/codebuild-ci.yml

@@ -55,7 +55,7 @@ jobs:
       - name: Run Codestyle & Doc Tests
         uses: aws-actions/aws-codebuild-run-build@v1
         with:
-          project-name: sagemaker-python-sdk-ci-codestyle-doc-tests
+          project-name: ${{ github.event.repository.name }}-ci-codestyle-doc-tests
           source-version-override: 'refs/pull/${{ github.event.pull_request.number }}/head^{${{ github.event.pull_request.head.sha }}}'
   unit-tests:
     runs-on: ubuntu-latest
@@ -74,7 +74,7 @@ jobs:
       - name: Run Unit Tests
         uses: aws-actions/aws-codebuild-run-build@v1
         with:
-          project-name: sagemaker-python-sdk-ci-unit-tests
+          project-name: ${{ github.event.repository.name }}-ci-unit-tests
           source-version-override: 'refs/pull/${{ github.event.pull_request.number }}/head^{${{ github.event.pull_request.head.sha }}}'
           env-vars-for-codebuild: |
             PY_VERSION
@@ -93,5 +93,5 @@ jobs:
       - name: Run Integ Tests
         uses: aws-actions/aws-codebuild-run-build@v1
         with:
-          project-name: sagemaker-python-sdk-ci-integ-tests
+          project-name: ${{ github.event.repository.name }}-ci-integ-tests
           source-version-override: 'refs/pull/${{ github.event.pull_request.number }}/head^{${{ github.event.pull_request.head.sha }}}'

CHANGELOG.md

@@ -1,5 +1,115 @@
 # Changelog
 
+## v2.222.1 (2024-06-12)
+
+### Bug Fixes and Other Changes
+
+ * First changes
+ * estimator.deploy not respecting instance type
+
+## v2.222.0 (2024-06-07)
+
+### Features
+
+ * jumpstart telemetry
+
+### Bug Fixes and Other Changes
+
+ * update image_uri_configs  06-06-2024 07:17:31 PST
+ * bump requests from 2.31.0 to 2.32.2 in /requirements/extras
+ * chore: add HF LLM neuronx 0.0.23 image
+ * Updates for DJL 0.28.0 release
+ * chore(deps): bump mlflow from 2.11.1 to 2.12.1 in /tests/data/serve_resources/mlflow/tensorflow
+ * chore(deps): bump mlflow from 2.11.1 to 2.12.1 in /tests/data/serve_resources/mlflow/xgboost
+ * chore(deps): bump mlflow from 2.10.2 to 2.12.1 in /tests/data/serve_resources/mlflow/pytorch
+ * chore(deps): bump apache-airflow from 2.9.0 to 2.9.1 in /requirements/extras
+ * chore(deps): bump requests from 2.31.0 to 2.32.2 in /tests/data/serve_resources/mlflow/pytorch
+ * Fix ci unit-tests
+ * Making project name in workflow files dynamic
+ * update image_uri_configs  05-29-2024 07:17:35 PST
+ * Update: SM Endpoint Routing Strategy Support.
+
+## v2.221.1 (2024-05-22)
+
+### Bug Fixes and Other Changes
+
+ * Convert pytorchddp distribution to smdistributed distribution
+ * Add tei cpu image
+
+## v2.221.0 (2024-05-20)
+
+### Features
+
+ * onboard tei image config to pysdk
+
+### Bug Fixes and Other Changes
+
+ * JS Model with non-TGI/non-DJL deployment failure
+ * cover tei with image_uris.retrieve API
+ * Add more debuging
+ * model builder limited container support for endpoint mode.
+ * Image URI should take precedence for HF models
+
+## v2.220.0 (2024-05-15)
+
+### Features
+
+ * AutoGluon 1.1.0 image_uris update
+ * add new images for HF TGI release
+ * Add telemetry support for mlflow models
+
+### Bug Fixes and Other Changes
+
+ * add debug logs to workflow container dist creation
+ * model builder race condition on sagemaker session
+ * Add tensorflow_serving support for mlflow models and enable lineage tracking for mlflow models
+ * update image_uri_configs  05-09-2024 07:17:41 PST
+ * skip flakey tests pending investigation
+
+## v2.219.0 (2024-05-08)
+
+### Features
+
+ * allow choosing js payload by alias in private method
+
+### Bug Fixes and Other Changes
+
+ * chore(deps): bump jinja2 from 3.1.3 to 3.1.4 in /requirements/extras
+ * chore(deps): bump tqdm from 4.66.2 to 4.66.3 in /tests/data/serve_resources/mlflow/pytorch
+ * chore(deps): bump jinja2 from 3.1.3 to 3.1.4 in /doc
+ * Updates for SMP v2.3.1
+
+## v2.218.1 (2024-05-03)
+
+### Bug Fixes and Other Changes
+
+ * Fix UserAgent logging in Python SDK
+ * chore: release tgi 2.0.1
+ * chore: update skipped flaky tests
+
+## v2.218.0 (2024-05-01)
+
+### Features
+
+ * set default allow_pickle param to False
+
+### Bug Fixes and Other Changes
+
+ * properly close files in lineage queries and tests
+
+## v2.217.0 (2024-04-24)
+
+### Features
+
+ * support session tag chaining for training job
+
+### Bug Fixes and Other Changes
+
+ * Add Triton v24.03 URI
+ * mainline alt config parsing
+ * Fix tox installs
+ * Add PT 2.2 Graviton Inference DLC
+
 ## v2.216.1 (2024-04-22)
 
 ### Bug Fixes and Other Changes

README.rst

@@ -95,6 +95,13 @@ SageMaker Python SDK is tested on:
 - Python 3.10
 - Python 3.11
 
+Telemetry
+~~~~~~~~~~~~~~~
+
+The ``sagemaker`` library has telemetry enabled to help us better understand user needs, diagnose issues, and deliver new features. This telemetry tracks the usage of various SageMaker functions.
+
+If you prefer to opt out of telemetry, you can easily do so by setting the ``TelemetryOptOut`` parameter to ``true`` in the SDK defaults configuration. For detailed instructions, please visit `Configuring and using defaults with the SageMaker Python SDK <https://sagemaker.readthedocs.io/en/stable/overview.html#configuring-and-using-defaults-with-the-sagemaker-python-sdk>`__.
+
 AWS Permissions
 ~~~~~~~~~~~~~~~
 

VERSION

@@ -1 +1 @@
-2.216.2.dev0
+2.222.2.dev0

doc/requirements.txt

@@ -2,6 +2,6 @@ sphinx==5.1.1
 sphinx-rtd-theme==0.5.0
 docutils==0.15.2
 packaging==20.9
-jinja2==3.1.3
+jinja2==3.1.4
 schema==0.7.5
 accelerate>=0.24.1,<=0.27.0

requirements/extras/test_requirements.txt

@@ -12,13 +12,13 @@ awslogs==0.14.0
 black==24.3.0
 stopit==1.1.2
 # Update tox.ini to have correct version of airflow constraints file
-apache-airflow==2.9.0
+apache-airflow==2.9.1
 apache-airflow-providers-amazon==7.2.1
 attrs>=23.1.0,<24
 fabric==2.6.0
-requests==2.31.0
+requests==2.32.2
 sagemaker-experiments==0.1.35
-Jinja2==3.1.3
+Jinja2==3.1.4
 pyvis==0.2.1
 pandas>=1.3.5,<1.5
 scikit-learn==1.3.0
@@ -36,3 +36,4 @@ onnx>=1.15.0
 nbformat>=5.9,<6
 accelerate>=0.24.1,<=0.27.0
 schema==0.7.5
+tensorflow>=2.1,<=2.16

src/sagemaker/base_deserializers.py

@@ -196,14 +196,14 @@ class NumpyDeserializer(SimpleBaseDeserializer):
     single array.
     """
 
-    def __init__(self, dtype=None, accept="application/x-npy", allow_pickle=True):
+    def __init__(self, dtype=None, accept="application/x-npy", allow_pickle=False):
         """Initialize a ``NumpyDeserializer`` instance.
 
         Args:
             dtype (str): The dtype of the data (default: None).
             accept (union[str, tuple[str]]): The MIME type (or tuple of allowable MIME types) that
                 is expected from the inference endpoint (default: "application/x-npy").
-            allow_pickle (bool): Allow loading pickled object arrays (default: True).
+            allow_pickle (bool): Allow loading pickled object arrays (default: False).
         """
         super(NumpyDeserializer, self).__init__(accept=accept)
         self.dtype = dtype
@@ -227,10 +227,21 @@ def deserialize(self, stream, content_type):
             if content_type == "application/json":
                 return np.array(json.load(codecs.getreader("utf-8")(stream)), dtype=self.dtype)
             if content_type == "application/x-npy":
-                return np.load(io.BytesIO(stream.read()), allow_pickle=self.allow_pickle)
+                try:
+                    return np.load(io.BytesIO(stream.read()), allow_pickle=self.allow_pickle)
+                except ValueError as ve:
+                    raise ValueError(
+                        "Please set the param allow_pickle=True \
+                        to deserialize pickle objects in NumpyDeserializer"
+                    ).with_traceback(ve.__traceback__)
             if content_type == "application/x-npz":
                 try:
                     return np.load(io.BytesIO(stream.read()), allow_pickle=self.allow_pickle)
+                except ValueError as ve:
+                    raise ValueError(
+                        "Please set the param allow_pickle=True \
+                        to deserialize pickle objectsin NumpyDeserializer"
+                    ).with_traceback(ve.__traceback__)
                 finally:
                     stream.close()
         finally:

src/sagemaker/chainer/model.py

@@ -28,6 +28,10 @@
 from sagemaker.model import FrameworkModel, MODEL_SERVER_WORKERS_PARAM_NAME
 from sagemaker.chainer import defaults
 from sagemaker.deserializers import NumpyDeserializer
+from sagemaker.model_card import (
+    ModelCard,
+    ModelPackageModelCard,
+)
 from sagemaker.predictor import Predictor
 from sagemaker.serializers import NumpySerializer
 from sagemaker.utils import to_string
@@ -175,6 +179,7 @@ def register(
         data_input_configuration: Optional[Union[str, PipelineVariable]] = None,
         skip_model_validation: Optional[Union[str, PipelineVariable]] = None,
         source_uri: Optional[Union[str, PipelineVariable]] = None,
+        model_card: Optional[Union[ModelPackageModelCard, ModelCard]] = None,
     ):
         """Creates a model package for creating SageMaker models or listing on Marketplace.
 
@@ -226,6 +231,8 @@ def register(
                 validation. Values can be "All" or "None" (default: None).
             source_uri (str or PipelineVariable): The URI of the source for the model package
                 (default: None).
+            model_card (ModeCard or ModelPackageModelCard): document contains qualitative and
+                quantitative information about a model (default: None).
 
         Returns:
             str: A string of SageMaker Model Package ARN.
@@ -266,6 +273,7 @@ def register(
             data_input_configuration=data_input_configuration,
             skip_model_validation=skip_model_validation,
             source_uri=source_uri,
+            model_card=model_card,
         )
 
     def prepare_container_def(

src/sagemaker/enums.py

@@ -28,3 +28,15 @@ class EndpointType(Enum):
     INFERENCE_COMPONENT_BASED = (
         "InferenceComponentBased"  # Amazon SageMaker Inference Component Based Endpoint
     )
+
+
+class RoutingStrategy(Enum):
+    """Strategy for routing https traffics."""
+
+    RANDOM = "RANDOM"
+    """The endpoint routes each request to a randomly chosen instance.
+    """
+    LEAST_OUTSTANDING_REQUESTS = "LEAST_OUTSTANDING_REQUESTS"
+    """The endpoint routes requests to the specific instances that have
+    more capacity to process them.
+    """

src/sagemaker/estimator.py

@@ -181,6 +181,7 @@ def __init__(
         container_arguments: Optional[List[str]] = None,
         disable_output_compression: bool = False,
         enable_remote_debug: Optional[Union[bool, PipelineVariable]] = None,
+        enable_session_tag_chaining: Optional[Union[bool, PipelineVariable]] = None,
         **kwargs,
     ):
         """Initialize an ``EstimatorBase`` instance.
@@ -544,7 +545,9 @@ def __init__(
             enable_infra_check (bool or PipelineVariable): Optional.
                 Specifies whether it is running Sagemaker built-in infra check jobs.
             enable_remote_debug (bool or PipelineVariable): Optional.
-                Specifies whether RemoteDebug is enabled for the training job
+                Specifies whether RemoteDebug is enabled for the training job.
+            enable_session_tag_chaining (bool or PipelineVariable): Optional.
+                Specifies whether SessionTagChaining is enabled for the training job.
         """
         instance_count = renamed_kwargs(
             "train_instance_count", "instance_count", instance_count, kwargs
@@ -785,6 +788,8 @@ def __init__(
 
         self._enable_remote_debug = enable_remote_debug
 
+        self._enable_session_tag_chaining = enable_session_tag_chaining
+
     @abstractmethod
     def training_image_uri(self):
         """Return the Docker image to use for training.
@@ -1719,6 +1724,7 @@ def register(
         data_input_configuration=None,
         skip_model_validation=None,
         source_uri=None,
+        model_card=None,
         **kwargs,
     ):
         """Creates a model package for creating SageMaker models or listing on Marketplace.
@@ -1767,6 +1773,8 @@ def register(
             skip_model_validation (str): Indicates if you want to skip model validation.
                 Values can be "All" or "None" (default: None).
             source_uri (str): The URI of the source for the model package (default: None).
+            model_card (ModeCard or ModelPackageModelCard): document contains qualitative and
+                quantitative information about a model (default: None).
             **kwargs: Passed to invocation of ``create_model()``. Implementations may customize
                 ``create_model()`` to accept ``**kwargs`` to customize model creation during
                 deploy. For more, see the implementation docs.
@@ -1812,6 +1820,7 @@ def register(
             data_input_configuration=data_input_configuration,
             skip_model_validation=skip_model_validation,
             source_uri=source_uri,
+            model_card=model_card,
         )
 
     @property
@@ -2318,6 +2327,14 @@ def get_remote_debug_config(self):
             else {"EnableRemoteDebug": self._enable_remote_debug}
         )
 
+    def get_session_chaining_config(self):
+        """dict: Return the configuration of SessionChaining"""
+        return (
+            None
+            if self._enable_session_tag_chaining is None
+            else {"EnableSessionTagChaining": self._enable_session_tag_chaining}
+        )
+
     def enable_remote_debug(self):
         """Enable remote debug for a training job."""
         self._update_remote_debug(True)
@@ -2574,6 +2591,9 @@ def _get_train_args(cls, estimator, inputs, experiment_config):
         if estimator.get_remote_debug_config() is not None:
             train_args["remote_debug_config"] = estimator.get_remote_debug_config()
 
+        if estimator.get_session_chaining_config() is not None:
+            train_args["session_chaining_config"] = estimator.get_session_chaining_config()
+
         return train_args
 
     @classmethod
@@ -2766,6 +2786,7 @@ def __init__(
         disable_output_compression: bool = False,
         enable_infra_check: Optional[Union[bool, PipelineVariable]] = None,
         enable_remote_debug: Optional[Union[bool, PipelineVariable]] = None,
+        enable_session_tag_chaining: Optional[Union[bool, PipelineVariable]] = None,
         **kwargs,
     ):
         """Initialize an ``Estimator`` instance.
@@ -3129,6 +3150,8 @@ def __init__(
                 Specifies whether it is running Sagemaker built-in infra check jobs.
             enable_remote_debug (bool or PipelineVariable): Optional.
                 Specifies whether RemoteDebug is enabled for the training job
+            enable_session_tag_chaining (bool or PipelineVariable): Optional.
+                 Specifies whether SessionTagChaining is enabled for the training job
         """
         self.image_uri = image_uri
         self._hyperparameters = hyperparameters.copy() if hyperparameters else {}
@@ -3181,6 +3204,7 @@ def __init__(
             container_arguments=container_arguments,
             disable_output_compression=disable_output_compression,
             enable_remote_debug=enable_remote_debug,
+            enable_session_tag_chaining=enable_session_tag_chaining,
             **kwargs,
         )