Sagemaker

Latest version: v2.223.0

Safety actively analyzes 638819 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 26 of 100

2.99.0

Not secure
Features

* heterogeneous cluster set up in distribution config
* support heterogeneous cluster for training
* include fields to work with inference recommender

Bug Fixes and Other Changes

* Moving the newly added field instance_group to the end of method
* image_uri does not need to be specified with instance_groups
* Loosen version of attrs dependency
* Add PipelineVariable annotation in estimatory, processing, tuner, transformer base classes
* model table link

Documentation Changes

* documentation for heterogeneous cluster

2.98.0

Not secure
Features

* Adding deepar image

Documentation Changes

* edit to clarify how to use inference.py

2.97.0

Not secure
Deprecations and Removals

* remove support for python 3.6

Features

* update prebuilt models documentation

Bug Fixes and Other Changes

* Skipping test_candidate_estimator_default_rerun_and_deploy
* Update model name from 'compiled.pt' to 'model.pth' for neo
* update pytest, skip hf integ temp
* Add override_pipeline_parameter_var decorator to give grace period to update invalid pipeline var args

2.96.0

Not secure
Features

* Add helper method to generate pipeline adjacency list

Bug Fixes and Other Changes

* changing trcomp integ tests to be able to run in all regions

2.95.0

Not secure
Features

* Adding Training Compiler support for TensorFlow estimator starting TF 2.9
* Add support for TF 2.9 training

Bug Fixes and Other Changes

* integs fallback from p3 to p2 instance
* bucket exists check for session.default_bucket
* make instance type fields as optional

Documentation Changes

* improvements on the docstring of ModelStep
* Add XGBoostProcessor

2.94.0

Not secure
Features

* AutoGluon 0.4.2 image_uris support

Page 26 of 100

Links

Releases

Has known vulnerabilities

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.