Merge branch 'master' into approved-mpg

Author: keshav-chandak

CHANGELOG.md

@@ -1,5 +1,21 @@
 # Changelog
 
+## v2.242.0 (2025-03-14)
+
+### Features
+
+ * add integ tests for training JumpStart models in private hub
+
+### Bug Fixes and Other Changes
+
+ * Torch upgrade
+ * Prevent RunContext overlap between test_run tests
+ * remove s3 output location requirement from hub class init
+ * Fixing Pytorch training python version in tests
+ * update image_uri_configs  03-11-2025 07:18:09 PST
+ * resolve infinite loop in _find_config on Windows systems
+ * pipeline definition function doc update
+
 ## v2.241.0 (2025-03-06)
 
 ### Features

VERSION

@@ -1 +1 @@
-2.241.1.dev0
+2.242.1.dev0

doc/overview.rst

@@ -30,6 +30,11 @@ To train a model by using the SageMaker Python SDK, you:
 
 After you train a model, you can save it, and then serve the model as an endpoint to get real-time inferences or get inferences for an entire dataset by using batch transform.
 
+
+Important Note:
+
+*  When using torch to load Models, it is recommended to use version torch>=2.6.0 and torchvision>=0.17.0
+
 Prepare a Training script
 =========================
 

src/sagemaker/huggingface/model.py

@@ -218,6 +218,7 @@ def deploy(
         container_startup_health_check_timeout=None,
         inference_recommendation_id=None,
         explainer_config=None,
+        update_endpoint: Optional[bool] = False,
         **kwargs,
     ):
         """Deploy this ``Model`` to an ``Endpoint`` and optionally return a ``Predictor``.
@@ -296,6 +297,11 @@ def deploy(
                 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)
+            update_endpoint (Optional[bool]):
+                Flag to update the model in an existing Amazon SageMaker endpoint.
+                If True, this will deploy a new EndpointConfig to an already existing endpoint
+                and delete resources corresponding to the previous EndpointConfig. Default: False
+                Note: Currently this is supported for single model endpoints
         Raises:
              ValueError: If arguments combination check failed in these circumstances:
                 - If no role is specified or
@@ -335,6 +341,7 @@ def deploy(
             container_startup_health_check_timeout=container_startup_health_check_timeout,
             inference_recommendation_id=inference_recommendation_id,
             explainer_config=explainer_config,
+            update_endpoint=update_endpoint,
             **kwargs,
         )
 

src/sagemaker/image_uri_config/pytorch.json

@@ -85,7 +85,8 @@
             "2.2": "2.2.0",
             "2.3": "2.3.0",
             "2.4": "2.4.0",
-            "2.5": "2.5.1"
+            "2.5": "2.5.1",
+            "2.6": "2.6.0"
         },
         "versions": {
             "0.4.0": {
@@ -1253,6 +1254,50 @@
                     "us-west-2": "763104351884"
                 },
                 "repository": "pytorch-inference"
+            },
+            "2.6.0": {
+                "py_versions": [
+                    "py312"
+                ],
+                "registries": {
+                    "af-south-1": "626614931356",
+                    "ap-east-1": "871362719292",
+                    "ap-northeast-1": "763104351884",
+                    "ap-northeast-2": "763104351884",
+                    "ap-northeast-3": "364406365360",
+                    "ap-south-1": "763104351884",
+                    "ap-south-2": "772153158452",
+                    "ap-southeast-1": "763104351884",
+                    "ap-southeast-2": "763104351884",
+                    "ap-southeast-3": "907027046896",
+                    "ap-southeast-4": "457447274322",
+                    "ap-southeast-5": "550225433462",
+                    "ap-southeast-7": "590183813437",
+                    "ca-central-1": "763104351884",
+                    "ca-west-1": "204538143572",
+                    "cn-north-1": "727897471807",
+                    "cn-northwest-1": "727897471807",
+                    "eu-central-1": "763104351884",
+                    "eu-central-2": "380420809688",
+                    "eu-north-1": "763104351884",
+                    "eu-south-1": "692866216735",
+                    "eu-south-2": "503227376785",
+                    "eu-west-1": "763104351884",
+                    "eu-west-2": "763104351884",
+                    "eu-west-3": "763104351884",
+                    "il-central-1": "780543022126",
+                    "me-central-1": "914824155844",
+                    "me-south-1": "217643126080",
+                    "mx-central-1": "637423239942",
+                    "sa-east-1": "763104351884",
+                    "us-east-1": "763104351884",
+                    "us-east-2": "763104351884",
+                    "us-gov-east-1": "446045086412",
+                    "us-gov-west-1": "442386744353",
+                    "us-west-1": "763104351884",
+                    "us-west-2": "763104351884"
+                },
+                "repository": "pytorch-inference"
             }
         }
     },
@@ -1628,7 +1673,8 @@
             "2.2": "2.2.0",
             "2.3": "2.3.0",
             "2.4": "2.4.0",
-            "2.5": "2.5.1"
+            "2.5": "2.5.1",
+            "2.6": "2.6.0"
         },
         "versions": {
             "0.4.0": {
@@ -2801,6 +2847,50 @@
                     "us-west-2": "763104351884"
                 },
                 "repository": "pytorch-training"
+            },
+            "2.6.0": {
+                "py_versions": [
+                    "py312"
+                ],
+                "registries": {
+                    "af-south-1": "626614931356",
+                    "ap-east-1": "871362719292",
+                    "ap-northeast-1": "763104351884",
+                    "ap-northeast-2": "763104351884",
+                    "ap-northeast-3": "364406365360",
+                    "ap-south-1": "763104351884",
+                    "ap-south-2": "772153158452",
+                    "ap-southeast-1": "763104351884",
+                    "ap-southeast-2": "763104351884",
+                    "ap-southeast-3": "907027046896",
+                    "ap-southeast-4": "457447274322",
+                    "ap-southeast-5": "550225433462",
+                    "ap-southeast-7": "590183813437",
+                    "ca-central-1": "763104351884",
+                    "ca-west-1": "204538143572",
+                    "cn-north-1": "727897471807",
+                    "cn-northwest-1": "727897471807",
+                    "eu-central-1": "763104351884",
+                    "eu-central-2": "380420809688",
+                    "eu-north-1": "763104351884",
+                    "eu-south-1": "692866216735",
+                    "eu-south-2": "503227376785",
+                    "eu-west-1": "763104351884",
+                    "eu-west-2": "763104351884",
+                    "eu-west-3": "763104351884",
+                    "il-central-1": "780543022126",
+                    "me-central-1": "914824155844",
+                    "me-south-1": "217643126080",
+                    "mx-central-1": "637423239942",
+                    "sa-east-1": "763104351884",
+                    "us-east-1": "763104351884",
+                    "us-east-2": "763104351884",
+                    "us-gov-east-1": "446045086412",
+                    "us-gov-west-1": "442386744353",
+                    "us-west-1": "763104351884",
+                    "us-west-2": "763104351884"
+                },
+                "repository": "pytorch-training"
             }
         }
     }

src/sagemaker/jumpstart/constants.py

@@ -81,6 +81,12 @@
             gated_content_bucket="jumpstart-private-cache-prod-eu-north-1",
             neo_content_bucket="sagemaker-sd-models-prod-eu-north-1",
         ),
+        JumpStartLaunchedRegionInfo(
+            region_name="eu-south-2",
+            content_bucket="jumpstart-cache-prod-eu-south-2",
+            gated_content_bucket="jumpstart-private-cache-prod-eu-south-2",
+            neo_content_bucket="sagemaker-sd-models-prod-eu-south-2",
+        ),
         JumpStartLaunchedRegionInfo(
             region_name="me-south-1",
             content_bucket="jumpstart-cache-prod-me-south-1",
@@ -97,6 +103,12 @@
             gated_content_bucket="jumpstart-private-cache-prod-ap-south-1",
             neo_content_bucket="sagemaker-sd-models-prod-ap-south-1",
         ),
+        JumpStartLaunchedRegionInfo(
+            region_name="ap-south-2",
+            content_bucket="jumpstart-cache-prod-ap-south-2",
+            gated_content_bucket="jumpstart-private-cache-prod-ap-south-2",
+            neo_content_bucket="sagemaker-sd-models-prod-ap-south-2",
+        ),
         JumpStartLaunchedRegionInfo(
             region_name="eu-west-3",
             content_bucket="jumpstart-cache-prod-eu-west-3",
@@ -137,6 +149,12 @@
             gated_content_bucket="jumpstart-private-cache-prod-ap-southeast-3",
             neo_content_bucket="sagemaker-sd-models-prod-ap-southeast-3",
         ),
+        JumpStartLaunchedRegionInfo(
+            region_name="ap-southeast-4",
+            content_bucket="jumpstart-cache-prod-ap-southeast-4",
+            gated_content_bucket="jumpstart-private-cache-prod-ap-southeast-4",
+            neo_content_bucket="sagemaker-sd-models-prod-ap-southeast-4",
+        ),
         JumpStartLaunchedRegionInfo(
             region_name="ap-southeast-5",
             content_bucket="jumpstart-cache-prod-ap-southeast-5",
@@ -188,6 +206,12 @@
             gated_content_bucket="jumpstart-private-cache-prod-ca-central-1",
             neo_content_bucket="sagemaker-sd-models-prod-ca-central-1",
         ),
+        JumpStartLaunchedRegionInfo(
+            region_name="ca-west-1",
+            content_bucket="jumpstart-cache-prod-ca-west-1",
+            gated_content_bucket="jumpstart-private-cache-prod-ca-west-1",
+            neo_content_bucket="sagemaker-sd-models-prod-ca-west-1",
+        ),
         JumpStartLaunchedRegionInfo(
             region_name="cn-north-1",
             content_bucket="jumpstart-cache-prod-cn-north-1",

src/sagemaker/jumpstart/hub/hub.py

@@ -16,35 +16,25 @@
 from datetime import datetime
 import logging
 from typing import Optional, Dict, List, Any, Union
-from botocore import exceptions
 
 from sagemaker.jumpstart.constants import JUMPSTART_MODEL_HUB_NAME
 from sagemaker.jumpstart.enums import JumpStartScriptScope
 from sagemaker.session import Session
 
-from sagemaker.jumpstart.constants import (
-    JUMPSTART_LOGGER,
-)
 from sagemaker.jumpstart.types import (
     HubContentType,
 )
 from sagemaker.jumpstart.filters import Constant, Operator, BooleanValues
 from sagemaker.jumpstart.hub.utils import (
     get_hub_model_version,
     get_info_from_hub_resource_arn,
-    create_hub_bucket_if_it_does_not_exist,
-    generate_default_hub_bucket_name,
-    create_s3_object_reference_from_uri,
     construct_hub_arn_from_name,
 )
 
 from sagemaker.jumpstart.notebook_utils import (
     list_jumpstart_models,
 )
 
-from sagemaker.jumpstart.hub.types import (
-    S3ObjectLocation,
-)
 from sagemaker.jumpstart.hub.interfaces import (
     DescribeHubResponse,
     DescribeHubContentResponse,
@@ -66,8 +56,8 @@ class Hub:
     def __init__(
         self,
         hub_name: str,
+        sagemaker_session: Session,
         bucket_name: Optional[str] = None,
-        sagemaker_session: Optional[Session] = None,
     ) -> None:
         """Instantiates a SageMaker ``Hub``.
 
@@ -78,41 +68,11 @@ def __init__(
         """
         self.hub_name = hub_name
         self.region = sagemaker_session.boto_region_name
+        self.bucket_name = bucket_name
         self._sagemaker_session = (
             sagemaker_session
             or utils.get_default_jumpstart_session_with_user_agent_suffix(is_hub_content=True)
         )
-        self.hub_storage_location = self._generate_hub_storage_location(bucket_name)
-
-    def _fetch_hub_bucket_name(self) -> str:
-        """Retrieves hub bucket name from Hub config if exists"""
-        try:
-            hub_response = self._sagemaker_session.describe_hub(hub_name=self.hub_name)
-            hub_output_location = hub_response["S3StorageConfig"].get("S3OutputPath")
-            if hub_output_location:
-                location = create_s3_object_reference_from_uri(hub_output_location)
-                return location.bucket
-            default_bucket_name = generate_default_hub_bucket_name(self._sagemaker_session)
-            JUMPSTART_LOGGER.warning(
-                "There is not a Hub bucket associated with %s. Using %s",
-                self.hub_name,
-                default_bucket_name,
-            )
-            return default_bucket_name
-        except exceptions.ClientError:
-            hub_bucket_name = generate_default_hub_bucket_name(self._sagemaker_session)
-            JUMPSTART_LOGGER.warning(
-                "There is not a Hub bucket associated with %s. Using %s",
-                self.hub_name,
-                hub_bucket_name,
-            )
-            return hub_bucket_name
-
-    def _generate_hub_storage_location(self, bucket_name: Optional[str] = None) -> None:
-        """Generates an ``S3ObjectLocation`` given a Hub name."""
-        hub_bucket_name = bucket_name or self._fetch_hub_bucket_name()
-        curr_timestamp = datetime.now().timestamp()
-        return S3ObjectLocation(bucket=hub_bucket_name, key=f"{self.hub_name}-{curr_timestamp}")
 
     def _get_latest_model_version(self, model_id: str) -> str:
         """Populates the lastest version of a model from specs no matter what is passed.
@@ -132,19 +92,22 @@ def create(
         tags: Optional[str] = None,
     ) -> Dict[str, str]:
         """Creates a hub with the given description"""
+        curr_timestamp = datetime.now().timestamp()
 
-        create_hub_bucket_if_it_does_not_exist(
-            self.hub_storage_location.bucket, self._sagemaker_session
-        )
+        request = {
+            "hub_name": self.hub_name,
+            "hub_description": description,
+            "hub_display_name": display_name,
+            "hub_search_keywords": search_keywords,
+            "tags": tags,
+        }
 
-        return self._sagemaker_session.create_hub(
-            hub_name=self.hub_name,
-            hub_description=description,
-            hub_display_name=display_name,
-            hub_search_keywords=search_keywords,
-            s3_storage_config={"S3OutputPath": self.hub_storage_location.get_uri()},
-            tags=tags,
-        )
+        if self.bucket_name:
+            request["s3_storage_config"] = {
+                "S3OutputPath": (f"s3://{self.bucket_name}/{self.hub_name}-{curr_timestamp}")
+            }
+
+        return self._sagemaker_session.create_hub(**request)
 
     def describe(self, hub_name: Optional[str] = None) -> DescribeHubResponse:
         """Returns descriptive information about the Hub"""