Skip to content

Commit 4f4fd80

Browse files
LokiiiiiiLokiiiiii
committed
Merge remote-tracking branch 'aws/master' into trcomp-pt-112
2 parents b1fdf5e + f1f0013 commit 4f4fd80

File tree

84 files changed

+7784
-191
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

84 files changed

+7784
-191
lines changed

‎CHANGELOG.md

+52
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,57 @@
11
# Changelog
22

3+
## v2.119.0 (2022-12-03)
4+
5+
### Features
6+
7+
* Add Code Owners file
8+
* Added transform with monitoring pipeline step in transformer
9+
* Update TF 2.9 and TF 2.10 inference DLCs
10+
* make estimator accept json file as modelparallel config
11+
* SageMaker Training Compiler does not support p4de instances
12+
* Add support for SparkML v3.3
13+
14+
### Bug Fixes and Other Changes
15+
16+
* Fix bug forcing uploaded tar to be named sourcedir
17+
* Update local_requirements.txt PyYAML version
18+
* refactoring : using with statement
19+
* Allow Py 3.7 for MMS Test Docker env
20+
* fix PySparkProcessor __init__ params type
21+
* type hint of PySparkProcessor __init__
22+
* Return ARM XGB/SKLearn tags if `image_scope` is `inference_graviton`
23+
* Update scipy to 1.7.3 to support M1 development envs
24+
* Fixing type hints for Spark processor that has instance type/count params in reverse order
25+
* Add DeepAR ap-northeast-3 repository.
26+
* Fix AsyncInferenceConfig documentation typo
27+
* fix ml_inf to ml_inf1 in Neo multi-version support
28+
* Fix type annotations
29+
* add neo mvp region accounts
30+
31+
## v2.118.0 (2022-12-01)
32+
33+
### Features
34+
35+
* Update boto3 version to 1.26.20
36+
* support table format option for create feature group.
37+
* Support Amazon SageMaker Model Cards
38+
* support monitoring alerts api
39+
* Support Amazon SageMaker AutoMLStep
40+
41+
### Bug Fixes and Other Changes
42+
43+
* integration test in anticipate of ProfilerConfig API changes
44+
* Add more integ test logic for AutoMLStep
45+
* update get_execution_role_arn to use role from DefaultSpaceSettings
46+
* bug on AutoMLInput to allow PipelineVariable
47+
* FinalMetricDataList is missing from the training job search resu…
48+
* add integration tests for Model Card
49+
* update AutoMLStep with cache improvement
50+
51+
### Documentation Changes
52+
53+
* automlstep doc update
54+
355
## v2.117.0 (2022-11-15)
456

557
### Features

‎CODEOWNERS

+1
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
* @aws/sagemaker-ml-frameworks

‎README.rst

+1-1
Original file line numberDiff line numberDiff line change
@@ -214,7 +214,7 @@ In order to host a SparkML model in SageMaker, it should be serialized with ``ML
214214

215215
For more information on MLeap, see https://github.com/combust/mleap .
216216

217-
Supported major version of Spark: 2.4 (MLeap version - 0.9.6)
217+
Supported major version of Spark: 3.3 (MLeap version - 0.20.0)
218218

219219
Here is an example on how to create an instance of ``SparkMLModel`` class and use ``deploy()`` method to create an
220220
endpoint which can be used to perform prediction against your trained SparkML Model.

‎VERSION

+1-1
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
2.117.1.dev0
1+
2.119.1.dev0

‎doc/amazon_sagemaker_model_building_pipeline.rst

+46
Original file line numberDiff line numberDiff line change
@@ -362,6 +362,52 @@ Example:
362362
prefix=model_prefix
363363
)
364364
365+
AutoMLStep
366+
`````````````
367+
Referable Property List:
368+
369+
- `DescribeAutoMLJob`_
370+
- `BestCandidateProperties.ModelInsightsJsonReportPath`_
371+
- `BestCandidateProperties.ExplainabilityJsonReportPath`_
372+
373+
.. _DescribeAutoMLJob: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJob
374+
.. _BestCandidateProperties.ModelInsightsJsonReportPath: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CandidateArtifactLocations.html#sagemaker-Type-CandidateArtifactLocations-ModelInsights
375+
.. _BestCandidateProperties.ExplainabilityJsonReportPath: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CandidateArtifactLocations.html#sagemaker-Type-CandidateArtifactLocations-Explainability
376+
377+
Example:
378+
379+
.. code-block:: python
380+
381+
step_automl = AutoMLStep(...)
382+
383+
auto_ml_model = step_automl.get_best_model(<role>)
384+
385+
model_metrics = ModelMetrics(
386+
model_statistics=MetricsSource(
387+
s3_uri=auto_ml_step.properties.BestCandidateProperties.ModelInsightsJsonReportPath,
388+
content_type="application/json",
389+
),
390+
explainability=MetricsSource(
391+
s3_uri=auto_ml_step.properties.BestCandidateProperties.ExplainabilityJsonReportPath,
392+
content_type="application/json",
393+
)
394+
)
395+
396+
step_args_register_model = auto_ml_model.register(
397+
content_types=["text/csv"],
398+
response_types=["text/csv"],
399+
inference_instances=["ml.t2.medium", "ml.m5.large"],
400+
transform_instances=["ml.m5.large"],
401+
model_package_group_name="My model package group name",
402+
approval_status="PendingManualApproval",
403+
model_metrics=model_metrics,
404+
)
405+
406+
step_register_model = ModelStep(
407+
name="auto-ml-model-register",
408+
step_args=step_args_register_model,
409+
)
410+
365411
CreateModelStep
366412
````````````````
367413
Referable Property List:

‎doc/api/governance/index.rst

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
###############
2+
Governance APIs
3+
###############
4+
5+
.. toctree::
6+
:maxdepth: 1
7+
:glob:
8+
9+
*

‎doc/api/governance/model_card.rst

+43
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,43 @@
1+
Amazon SageMaker Model Cards
2+
----------------------------
3+
4+
For more information about using model cards with the SageMaker Python SDK,
5+
see `Amazon SageMaker Model Cards <https://docs.aws.amazon.com/sagemaker/latest/dg/model-cards.html>`__ in the *Amazon SageMaker Developer Guide*.
6+
7+
.. currentmodule:: sagemaker.model_card
8+
9+
.. autoclass:: AdditionalInformation
10+
:show-inheritance:
11+
12+
.. autoclass:: Environment
13+
:show-inheritance:
14+
15+
.. autoclass:: EvaluationJob
16+
:show-inheritance:
17+
18+
.. autoclass:: Function
19+
:show-inheritance:
20+
21+
.. autoclass:: IntendedUses
22+
:show-inheritance:
23+
24+
.. autoclass:: Metric
25+
:show-inheritance:
26+
27+
.. autoclass:: MetricGroup
28+
:show-inheritance:
29+
30+
.. autoclass:: ModelCard
31+
:show-inheritance:
32+
33+
.. autoclass:: ModelOverview
34+
:show-inheritance:
35+
36+
.. autoclass:: ObjectiveFunction
37+
:show-inheritance:
38+
39+
.. autoclass:: TrainingDetails
40+
:show-inheritance:
41+
42+
.. autoclass:: TrainingJobDetails
43+
:show-inheritance:

‎doc/api/index.rst

+1
Original file line numberDiff line numberDiff line change
@@ -11,4 +11,5 @@ The SageMaker Python SDK consists of a variety classes for preparing data, train
1111
training/index
1212
training/distributed
1313
inference/index
14+
governance/index
1415
utility/index

‎doc/overview.rst

+2-2
Original file line numberDiff line numberDiff line change
@@ -1171,7 +1171,7 @@ be ``s3://sagemaker-{REGION}-{ACCOUNTID}/async-endpoint-outputs/{UNIQUE-JOB-NAME
11711171
from sagemaker.async_inference import AsyncInferenceConfig
11721172
11731173
# Create an empty AsyncInferenceConfig object to use default values
1174-
async_config = new AsyncInferenceConfig()
1174+
async_config = AsyncInferenceConfig()
11751175
11761176
Or you can specify configurations in ``AsyncInferenceConfig`` as you like. All of those configuration parameters
11771177
are optional but if you don’t specify the ``output_path``, Amazon SageMaker will use the default ``S3OutputPath``
@@ -1180,7 +1180,7 @@ mentioned above (example shown below):
11801180
.. code:: python
11811181
11821182
# Specify S3OutputPath, MaxConcurrentInvocationsPerInstance and NotificationConfig in the async config object
1183-
async_config = new AsyncInferenceConfig(
1183+
async_config = AsyncInferenceConfig(
11841184
output_path="s3://{s3_bucket}/{bucket_prefix}/output",
11851185
max_concurrent_invocations_per_instance=10,
11861186
notification_config = {

‎doc/workflows/pipelines/sagemaker.workflow.pipelines.rst

+2
Original file line numberDiff line numberDiff line change
@@ -168,3 +168,5 @@ Steps
168168
.. autoclass:: sagemaker.workflow.clarify_check_step.ClarifyCheckStep
169169

170170
.. autoclass:: sagemaker.workflow.fail_step.FailStep
171+
172+
.. autoclass:: sagemaker.workflow.automl_step.AutoMLStep

‎requirements/extras/scipy_requirements.txt

+1-1
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
scipy==1.7.2
1+
scipy==1.7.3

‎setup.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -48,7 +48,7 @@ def read_requirements(filename):
4848
# Declare minimal set for installation
4949
required_packages = [
5050
"attrs>=20.3.0,<23",
51-
"boto3>=1.20.21,<2.0",
51+
"boto3>=1.26.20,<2.0",
5252
"google-pasta",
5353
"numpy>=1.9.0,<2.0",
5454
"protobuf>=3.1,<4.0",

0 commit comments

Comments
 (0)