Merge master into master-assets

.github/workflows/codebuild-ci-health.yml

@@ -0,0 +1,97 @@
+name: CI Health
+on:
+  schedule:
+    - cron: "0 */3 * * *"
+  workflow_dispatch:
+
+permissions:
+    id-token: write # This is required for requesting the JWT
+
+jobs:
+  codestyle-doc-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Codestyle & Doc Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        with:
+          project-name: sagemaker-python-sdk-ci-health-codestyle-doc-tests
+  unit-tests:
+    runs-on: ubuntu-latest
+    strategy:
+        fail-fast: false
+        matrix:
+          python-version: ["py38", "py39", "py310"]
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Unit Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        with:
+          project-name: sagemaker-python-sdk-ci-health-unit-tests
+          env-vars-for-codebuild: |
+            PY_VERSION
+        env:
+          PY_VERSION: ${{ matrix.python-version }}
+  integ-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Integ Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        id: codebuild
+        with:
+          project-name: sagemaker-python-sdk-ci-health-integ-tests
+  slow-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Slow Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        with:
+          project-name: sagemaker-python-sdk-ci-health-slow-tests
+  localmode-tests:
+      runs-on: ubuntu-latest
+      steps:
+        - name: Configure AWS Credentials
+          uses: aws-actions/configure-aws-credentials@v4
+          with:
+            role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+            aws-region: us-west-2
+            role-duration-seconds: 10800
+        - name: Run Local Mode Tests
+          uses: aws-actions/aws-codebuild-run-build@v1
+          with:
+            project-name: sagemaker-python-sdk-ci-health-localmode-tests
+  notebook-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Notebook Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        with:
+          project-name: sagemaker-python-sdk-ci-health-notebook-tests

.github/workflows/codebuild-ci.yml

@@ -0,0 +1,48 @@
+name: PR Checks
+on:
+    pull_request_target:
+
+concurrency:
+    group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.head_ref }}
+    cancel-in-progress: true
+
+permissions:
+    id-token: write # This is required for requesting the JWT
+
+jobs:
+  codestyle-doc-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Codestyle & Doc Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        with:
+          project-name: sagemaker-python-sdk-ci-codestyle-doc-tests
+          source-version-override: 'pr/${{ github.event.pull_request.number }}'
+  unit-tests:
+    runs-on: ubuntu-latest
+    strategy:
+        fail-fast: false
+        matrix:
+          python-version: ["py38", "py39", "py310"]
+    steps:
+      - name: Configure AWS Credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.CI_AWS_ROLE_ARN }}
+          aws-region: us-west-2
+          role-duration-seconds: 10800
+      - name: Run Unit Tests
+        uses: aws-actions/aws-codebuild-run-build@v1
+        with:
+          project-name: sagemaker-python-sdk-ci-unit-tests
+          source-version-override: 'pr/${{ github.event.pull_request.number }}'
+          env-vars-for-codebuild: |
+            PY_VERSION
+        env:
+          PY_VERSION: ${{ matrix.python-version }}

CHANGELOG.md

@@ -1,5 +1,66 @@
 # Changelog
 
+## v2.213.0 (2024-03-15)
+
+### Features
+
+ * Add support for Streaming Inference
+ * tgi optimum 0.0.19, 0.0.20 releases
+ * support JumpStart proprietary models
+ * Add ModelDataSource and SourceUri support for model package and while registering
+ * Accept user-defined env variables for the entry-point
+ * Add overriding logic in ModelBuilder when task is provided
+
+### Bug Fixes and Other Changes
+
+ * Improvement of the tuner documentation
+ * Skip of tests which are long running and causing the ResourceLimitInUse exception
+ * Add AutoML -> AutoMLV2 mapper
+ * add ci-health checks
+ * split coverage out from testenv in tox.ini
+ * add PT 2.2 support for smdistributed, pytorchddp, and torch_distributed distributions
+ * sagemaker session region not being used
+ * chore: emit warning when no instance specific gated training env var is available, and raise exception when accept_eula flag is not supplied
+ * enable github actions for PRs
+ * Move sagemaker pysdk version check after bootstrap in remote job
+ * make unit tests compatible with pytest-xdist
+ * Update tblib constraint
+
+## v2.212.0 (2024-03-06)
+
+### Features
+
+ * Update SM Python SDK for PT 2.2.0 SM DLC
+
+### Bug Fixes and Other Changes
+
+ * Create custom tarfile extractall util to fix backward compatibility issue
+ * Upgrade smp to version 2.2
+ * Enhance model builder selection logic to include model size
+
+## v2.211.0 (2024-03-05)
+
+### Features
+
+ * pin dll version to support python3.11 to the sdk
+ * instance specific jumpstart host requirements
+ * Add TensorFlow 2.14 image configs
+ * Add AutoMLV2 support
+ * Support selective pipeline execution between function step and regular step
+ * Add new Triton DLC URIs
+
+### Bug Fixes and Other Changes
+
+ * Skip No Canvas regions for test_deploy_best_candidate
+ * make sure gpus are found in local_gpu run
+ * Bump Apache Airflow version to 2.8.2
+ * properly close sagemaker config file after loading config
+ * remove enable_network_isolation from the python doc
+
+### Documentation Changes
+
+ * Add doc for new feature processor APIs and classes
+
 ## v2.210.0 (2024-02-28)
 
 ### Features

CONTRIBUTING.md

@@ -77,8 +77,8 @@ Before sending us a pull request, please ensure that:
 1. Install coverage using `pip install .[test]`
 1. cd into the sagemaker-python-sdk folder: `cd sagemaker-python-sdk` or `cd /environment/sagemaker-python-sdk`
 1. Run the following tox command and verify that all code checks and unit tests pass: `tox tests/unit`
-
-You can also run a single test with the following command: `tox -e py310 -- -s -vv <path_to_file><file_name>::<test_function_name>`
+1. You can also run a single test with the following command: `tox -e py310 -- -s -vv <path_to_file><file_name>::<test_function_name>`
+1. You can run coverage via runcvoerage env : `tox -e runcoverage -- tests/unit` or `tox -e py310 -- tests/unit --cov=sagemaker --cov-append --cov-report xml`
   * Note that the coverage test will fail if you only run a single test, so make sure to surround the command with `export IGNORE_COVERAGE=-` and `unset IGNORE_COVERAGE`
   * Example: `export IGNORE_COVERAGE=- ; tox -e py310 -- -s -vv tests/unit/test_estimator.py::test_sagemaker_model_s3_uri_invalid ; unset IGNORE_COVERAGE`
 

README.rst

@@ -22,6 +22,10 @@ SageMaker Python SDK
    :target: https://sagemaker.readthedocs.io/en/stable/
    :alt: Documentation Status
 
+.. image:: https://github.com/benieric/sagemaker-python-sdk/actions/workflows/codebuild-ci-health.yml/badge.svg
+    :target: https://github.com/benieric/sagemaker-python-sdk/actions/workflows/codebuild-ci-health.yml
+    :alt: CI Health
+
 SageMaker Python SDK is an open source library for training and deploying machine learning models on Amazon SageMaker.
 
 With the SDK, you can train and deploy models using popular deep learning frameworks **Apache MXNet** and **TensorFlow**.
@@ -63,11 +67,10 @@ Table of Contents
 Installing the SageMaker Python SDK
 -----------------------------------
 
-The SageMaker Python SDK is built to PyPI and can be installed with pip as follows:
-
+The SageMaker Python SDK is built to PyPI and the latest version of the SageMaker Python SDK can be installed with pip as follows
 ::
 
-    pip install sagemaker
+    pip install sagemaker==<Latest version from pyPI from https://pypi.org/project/sagemaker/>
 
 You can install from source by cloning this repository and running a pip install command in the root directory of the repository:
 

VERSION

@@ -1 +1 @@
-2.210.1.dev0
+2.213.1.dev0

doc/doc_utils/jumpstart_doc_utils.py

@@ -74,9 +74,12 @@ class Frameworks(str, Enum):
 
 JUMPSTART_REGION = "eu-west-2"
 SDK_MANIFEST_FILE = "models_manifest.json"
+PROPRIETARY_SDK_MANIFEST_FILE = "proprietary-sdk-manifest.json"
 JUMPSTART_BUCKET_BASE_URL = "https://jumpstart-cache-prod-{}.s3.{}.amazonaws.com".format(
     JUMPSTART_REGION, JUMPSTART_REGION
 )
+PROPRIETARY_DOC_BUCKET = "https://jumpstart-cache-prod-us-west-2.s3.us-west-2.amazonaws.com"
+
 TASK_MAP = {
     Tasks.IC: ProblemTypes.IMAGE_CLASSIFICATION,
     Tasks.IC_EMBEDDING: ProblemTypes.IMAGE_EMBEDDING,
@@ -152,18 +155,26 @@ class Frameworks(str, Enum):
 }
 
 
-def get_jumpstart_sdk_manifest():
-    url = "{}/{}".format(JUMPSTART_BUCKET_BASE_URL, SDK_MANIFEST_FILE)
+def get_public_s3_json_object(url):
     with request.urlopen(url) as f:
         models_manifest = f.read().decode("utf-8")
     return json.loads(models_manifest)
 
 
-def get_jumpstart_sdk_spec(key):
-    url = "{}/{}".format(JUMPSTART_BUCKET_BASE_URL, key)
-    with request.urlopen(url) as f:
-        model_spec = f.read().decode("utf-8")
-    return json.loads(model_spec)
+def get_jumpstart_sdk_manifest():
+    return get_public_s3_json_object(f"{JUMPSTART_BUCKET_BASE_URL}/{SDK_MANIFEST_FILE}")
+
+
+def get_proprietary_sdk_manifest():
+    return get_public_s3_json_object(f"{PROPRIETARY_DOC_BUCKET}/{PROPRIETARY_SDK_MANIFEST_FILE}")
+
+
+def get_jumpstart_sdk_spec(s3_key: str):
+    return get_public_s3_json_object(f"{JUMPSTART_BUCKET_BASE_URL}/{s3_key}")
+
+
+def get_proprietary_sdk_spec(s3_key: str):
+    return get_public_s3_json_object(f"{PROPRIETARY_DOC_BUCKET}/{s3_key}")
 
 
 def get_model_task(id):
@@ -196,6 +207,45 @@ def get_model_source(url):
         return "Source"
 
 
+def create_proprietary_model_table():
+    proprietary_content_intro = []
+    proprietary_content_intro.append("\n")
+    proprietary_content_intro.append(".. list-table:: Available Proprietary Models\n")
+    proprietary_content_intro.append("   :widths: 50 20 20 20 20\n")
+    proprietary_content_intro.append("   :header-rows: 1\n")
+    proprietary_content_intro.append("   :class: datatable\n")
+    proprietary_content_intro.append("\n")
+    proprietary_content_intro.append("   * - Model ID\n")
+    proprietary_content_intro.append("     - Fine Tunable?\n")
+    proprietary_content_intro.append("     - Supported Version\n")
+    proprietary_content_intro.append("     - Min SDK Version\n")
+    proprietary_content_intro.append("     - Source\n")
+
+    sdk_manifest = get_proprietary_sdk_manifest()
+    sdk_manifest_top_versions_for_models = {}
+
+    for model in sdk_manifest:
+        if model["model_id"] not in sdk_manifest_top_versions_for_models:
+            sdk_manifest_top_versions_for_models[model["model_id"]] = model
+        else:
+            if str(sdk_manifest_top_versions_for_models[model["model_id"]]["version"]) < str(
+                model["version"]
+            ):
+                sdk_manifest_top_versions_for_models[model["model_id"]] = model
+
+    proprietary_content_entries = []
+    for model in sdk_manifest_top_versions_for_models.values():
+        model_spec = get_proprietary_sdk_spec(model["spec_key"])
+        proprietary_content_entries.append("   * - {}\n".format(model_spec["model_id"]))
+        proprietary_content_entries.append("     - {}\n".format(False))  # TODO: support training
+        proprietary_content_entries.append("     - {}\n".format(model["version"]))
+        proprietary_content_entries.append("     - {}\n".format(model["min_version"]))
+        proprietary_content_entries.append(
+            "     - `{} <{}>`__ |external-link|\n".format("Source", model_spec.get("url"))
+        )
+    return proprietary_content_intro + proprietary_content_entries + ["\n"]
+
+
 def create_jumpstart_model_table():
     sdk_manifest = get_jumpstart_sdk_manifest()
     sdk_manifest_top_versions_for_models = {}
@@ -249,19 +299,19 @@ def create_jumpstart_model_table():
     file_content_intro.append("     - Source\n")
 
     dynamic_table_files = []
-    file_content_entries = []
+    open_weight_content_entries = []
 
     for model in sdk_manifest_top_versions_for_models.values():
         model_spec = get_jumpstart_sdk_spec(model["spec_key"])
         model_task = get_model_task(model_spec["model_id"])
         string_model_task = get_string_model_task(model_spec["model_id"])
         model_source = get_model_source(model_spec["url"])
-        file_content_entries.append("   * - {}\n".format(model_spec["model_id"]))
-        file_content_entries.append("     - {}\n".format(model_spec["training_supported"]))
-        file_content_entries.append("     - {}\n".format(model["version"]))
-        file_content_entries.append("     - {}\n".format(model["min_version"]))
-        file_content_entries.append("     - {}\n".format(model_task))
-        file_content_entries.append(
+        open_weight_content_entries.append("   * - {}\n".format(model_spec["model_id"]))
+        open_weight_content_entries.append("     - {}\n".format(model_spec["training_supported"]))
+        open_weight_content_entries.append("     - {}\n".format(model["version"]))
+        open_weight_content_entries.append("     - {}\n".format(model["min_version"]))
+        open_weight_content_entries.append("     - {}\n".format(model_task))
+        open_weight_content_entries.append(
             "     - `{} <{}>`__ |external-link|\n".format(model_source, model_spec["url"])
         )
 
@@ -299,7 +349,10 @@ def create_jumpstart_model_table():
             f.writelines(file_content_single_entry)
             f.close()
 
+    proprietary_content_entries = create_proprietary_model_table()
+
     f = open("doc_utils/pretrainedmodels.rst", "a")
     f.writelines(file_content_intro)
-    f.writelines(file_content_entries)
+    f.writelines(open_weight_content_entries)
+    f.writelines(proprietary_content_entries)
     f.close()

doc/requirements.txt

@@ -4,3 +4,4 @@ docutils==0.15.2
 packaging==20.9
 jinja2==3.1.3
 schema==0.7.5
+accelerate>=0.24.1,<=0.27.0

requirements/extras/huggingface_requirements.txt

@@ -0,0 +1 @@
+accelerate>=0.24.1,<=0.27.0

requirements/extras/test_requirements.txt

@@ -39,3 +39,4 @@ tritonclient[http]<2.37.0
 onnx==1.14.1
 # tf2onnx==1.15.1
 nbformat>=5.9,<6
+accelerate>=0.24.1,<=0.27.0