Merge branch 'master' into framework-processor-custom-entrypoint

Author: aviruthen

.gitignore

@@ -42,3 +42,4 @@ sagemaker_train/src/**/container_drivers/sourcecode.json
 sagemaker_train/src/**/container_drivers/distributed.json
 .kiro
 docs/api/generated/
+.hypothesis

.readthedocs.yaml

@@ -11,9 +11,7 @@ build:
 sphinx:
   configuration: docs/conf.py
 
-formats:
-  - pdf
-  - epub
+formats: []
 
 python:
   install:

CHANGELOG.md

@@ -1,4 +1,48 @@
 # Changelog
+## v3.10.0 (2026-05-01)
+
+### New Features
+- Make _PipelineExecution a public class
+- Add CodeArtifact support for ModelTrainer and FrameworkProcessor requirements.txt installation
+
+### Bug Fixes
+- Fix S3 bucket operations
+- Fix potential S3 path traversal
+- Wire FrameworkProcessor code_location into code upload paths
+- Improve subprocess exception handling in git_utils
+
+### Other
+- Update service-2.json with latest public botocore service model
+
+## v3.9.0 (2026-04-23)
+
+### New Features
+- **Train**: Add `wait_timeout` parameter to `train()` for SFT, DPO, RLAIF, RLVR, and BaseTrainer
+- **Evaluate**: Add MLflow experiment link to eval output
+- **JumpStart**: Allow `SAGEMAKER_HUB_NAME` environment variable to override the `HUB_NAME` constant
+
+### Bug Fixes
+- **HyperparameterTuner**: Pass through full `OutputDataConfig` from `ModelTrainer` so `kms_key_id`, `compression_type`, and other fields are preserved
+- **HyperparameterTuner / ModelTrainer**: Propagate environment variables that were previously dropped
+- **sagemaker-core**: Improve error messages for waiter timeouts
+- **ModelBuilder**: Stop overwriting user-provided `HF_MODEL_ID` for DJL Serving
+- **ModelBuilder**: Keep `/opt/ml/model` writable when using `source_code` with DJL LMI
+- **Evaluate**: Skip `None` hyperparameters in `to_dict` instead of converting them to the string `"None"`
+- **Nova**: Add `us-west-2` to Nova supported regions
+- **DJL LMI**: Update ISO account mappings
+
+## v3.8.0 (2026-04-16)
+
+### New Features
+- **Feature Group Manager**: Feature Group Manager support
+- **Image Upgrades**: Image upgrades
+
+### Bug Fixes
+- **ModelBuilder**: Add MLFlowConfig to Base Model
+- **Docker**: Support for docker compose > v2
+- **HuggingFace**: Improve SDK v3 Hugging Face support
+- **Dependencies**: Remove Pytorch hard dependency
+
 ## v3.7.1 (2026-03-31)
 
 ### Features

VERSION

@@ -1 +1 @@
-3.7.1
+3.10.0

docs/api/sagemaker_mlops.rst

@@ -21,19 +21,18 @@ Workflow Management
    :undoc-members:
    :show-inheritance:
 
-Local Development
------------------
+Feature Store
+-------------
 
-.. automodule:: sagemaker.mlops.local
+.. automodule:: sagemaker.mlops.feature_store
    :members:
    :undoc-members:
    :show-inheritance:
 
+Local Development
+-----------------
 
-Feature Store
--------------
-
-.. automodule:: sagemaker.mlops.feature_store
+.. automodule:: sagemaker.mlops.local
    :members:
    :undoc-members:
    :show-inheritance:

docs/conf.py

@@ -105,6 +105,25 @@
     'sagemaker.serve.model_server.torchserve.inference',
     'sagemaker.serve.model_server.torchserve.xgboost_inference',
     'sagemaker.serve.model_server.triton.model',
+    # Heavy runtime dependencies — mock to avoid slow/failing imports during doc build
+    'boto3',
+    'botocore',
+    'torch',
+    'tensorflow',
+    'sklearn',
+    'numpy',
+    'pandas',
+    'scipy',
+    'xgboost',
+    'lightgbm',
+    'mlflow',
+    'pyspark',
+    'transformers',
+    'datasets',
+    'trl',
 ]
 
 suppress_warnings = ['autodoc.import_error']
+
+# Speed up builds: only regenerate autosummary stubs if they don't exist
+autosummary_generate_overwrite = False

pyproject.toml

@@ -33,10 +33,10 @@ classifiers = [
   "Programming Language :: Python :: 3.12",
 ]
 dependencies = [
-  "sagemaker-core>=2.7.1,<3.0.0",
-  "sagemaker-train>=1.7.1,<2.0.0",
-  "sagemaker-serve>=1.7.1,<2.0.0",
-  "sagemaker-mlops>=1.7.1,<2.0.0",
+  "sagemaker-core>=2.10.0,<3.0.0",
+  "sagemaker-train>=1.10.0,<2.0.0",
+  "sagemaker-serve>=1.10.0,<2.0.0",
+  "sagemaker-mlops>=1.10.0,<2.0.0",
 ]
 
 [project.optional-dependencies]

requirements/extras/test_requirements.txt

@@ -12,4 +12,6 @@ omegaconf
 graphene
 typing_extensions>=4.9.0
 tensorflow>=2.16.2,<=2.19.0
-build
+build
+docker>=5.0.2,<8.0
+filelock>=3.0.0

sagemaker-core/CHANGELOG.md

@@ -1,4 +1,40 @@
 # Changelog
+## v2.10.0 (2026-05-01)
+
+### New Features
+
+- Add CodeArtifact support for ModelTrainer and FrameworkProcessor requirements.txt installation
+
+### Bug Fixes
+
+- Fix S3 bucket operations
+- Fix potential S3 path traversal
+- Wire FrameworkProcessor code_location into code upload paths
+- Improve subprocess exception handling in git_utils
+
+### Other
+
+- Update service-2.json with latest public botocore service model
+
+## v2.9.0 (2026-04-23)
+
+### Bug Fixes
+
+- **Waiters**: Improve error messages for waiter timeouts
+- **DJL LMI**: Update ISO account mappings in image URI config
+
+## v2.8.0 (2026-04-16)
+
+### New Features
+
+- **Image Upgrades**: Updated image URI configurations
+
+### Bug Fixes
+
+- **ModelBuilder**: Add MLFlowConfig to Base Model
+- **Local Mode**: Support for docker compose > v2
+- **Dependencies**: Remove Pytorch hard dependency
+
 ## v2.7.1 (2026-03-31)
 
 ### Features

sagemaker-core/VERSION

@@ -1 +1 @@
-2.7.1
+2.10.0