Merge branch 'main' into batch_tabular_model

* feat: notebook for AutoML tabular batch prediction

* feat: notebook for AutoML tabular batch prediction

Co-authored-by: gericdong <itseric@google.com>

.cloud-build/cleanup/cleanup-requirements.txt

@@ -1 +1,2 @@
+ratemate
 google-cloud-aiplatform

.cloud-build/cleanup/cleanup.py

@@ -1,9 +1,13 @@
 from typing import List
+from ratemate import RateLimit
+from resource_cleanup_manager import (
+    DatasetResourceCleanupManager,
+    ModelResourceCleanupManager,
+    EndpointResourceCleanupManager,
+    ResourceCleanupManager,
+)
 
-from resource_cleanup_manager import (DatasetResourceCleanupManager,
-                                      EndpointResourceCleanupManager,
-                                      ModelResourceCleanupManager,
-                                      ResourceCleanupManager)
+rate_limit = RateLimit(max_count=25, per=60, greedy=False)
 
 
 def run_cleanup_managers(managers: List[ResourceCleanupManager], is_dry_run: bool):
@@ -14,17 +18,18 @@ def run_cleanup_managers(managers: List[ResourceCleanupManager], is_dry_run: boo
         resources = manager.list()
         print(f"Found {len(resources)} {type_name}'s")
         for resource in resources:
-            if not manager.is_deletable(resource):
-                continue
+            try:
+                if not manager.is_deletable(resource):
+                    continue
 
-            if is_dry_run:
-                resource_name = manager.resource_name(resource)
-                print(f"Will delete '{type_name}': {resource_name}")
-            else:
-                try:
+                if is_dry_run:
+                    resource_name = manager.resource_name(resource)
+                    print(f"Will delete '{type_name}': {resource_name}")
+                else:
+                    rate_limit.wait()  # wait before deleting
                     manager.delete(resource)
-                except Exception as exception:
-                    print(exception)
+            except Exception as exception:
+                print(exception)
 
         print("")
 
@@ -38,7 +43,7 @@ if is_dry_run:
 managers = [
     DatasetResourceCleanupManager(),
     EndpointResourceCleanupManager(),
-    ModelResourceCleanupManager(),
+    ModelResourceCleanupManager(),  # ModelResourceCleanupManager must follow EndpointResourceCleanupManager due to deployed models blocking model deletion.
 ]
 
 run_cleanup_managers(managers=managers, is_dry_run=is_dry_run)

.cloud-build/cleanup/resource_cleanup_manager.py

@@ -1,5 +1,5 @@
 import abc
-from typing import Any
+from typing import Any, Type
 
 from google.cloud import aiplatform
 from google.cloud.aiplatform import base
@@ -41,7 +41,7 @@ class ResourceCleanupManager(abc.ABC):
         # Check that it wasn't created too recently, to prevent race conditions
         if time_difference <= RESOURCE_UPDATE_BUFFER_IN_SECONDS:
             print(
-                f"Skipping '{resource}' due update_time being '{time_difference}', which is less than '{RESOURCE_UPDATE_BUFFER_IN_SECONDS}'."
+                f"Skipping '{resource}' due to update_time being '{time_difference}', which is less than '{RESOURCE_UPDATE_BUFFER_IN_SECONDS}'."
             )
             return False
 
@@ -51,7 +51,7 @@ class ResourceCleanupManager(abc.ABC):
 class VertexAIResourceCleanupManager(ResourceCleanupManager):
     @property
     @abc.abstractmethod
-    def vertex_ai_resource(self) -> base.VertexAiResourceNounWithFutureManager:
+    def vertex_ai_resource(self) -> Type[base.VertexAiResourceNounWithFutureManager]:
         pass
 
     @property
@@ -61,7 +61,9 @@ class VertexAIResourceCleanupManager(ResourceCleanupManager):
     def list(self) -> Any:
         return self.vertex_ai_resource.list()
 
-    def resource_name(self, resource: Any) -> str:
+    def resource_name(
+        self, resource: Type[base.VertexAiResourceNounWithFutureManager]
+    ) -> str:
         return resource.display_name
 
     def delete(self, resource):
@@ -75,12 +77,33 @@ class VertexAIResourceCleanupManager(ResourceCleanupManager):
 
 class DatasetResourceCleanupManager(VertexAIResourceCleanupManager):
     vertex_ai_resource = aiplatform.datasets._Dataset
+    dataset_types = [
+        aiplatform.ImageDataset,
+        aiplatform.TabularDataset,
+        aiplatform.TextDataset,
+        aiplatform.TimeSeriesDataset,
+        aiplatform.VideoDataset,
+    ]
+
+    def list(self) -> Any:
+        return [
+            dataset
+            for dataset_type in self.dataset_types
+            for dataset in dataset_type.list()
+        ]
 
 
 class EndpointResourceCleanupManager(VertexAIResourceCleanupManager):
     vertex_ai_resource = aiplatform.Endpoint
 
     def delete(self, resource):
+        # TODO: Remove this once https://github.com/googleapis/python-aiplatform/issues/1441 is fixed
+        resource._sync_gca_resource()
+        for deployed_model_id in [
+            models.id for models in resource._gca_resource.deployed_models
+        ]:
+            resource._undeploy(deployed_model_id=deployed_model_id)
+
         resource.delete(force=True)
 
 

.cloud-build/execute_changed_notebooks_cli.py

@@ -62,6 +62,18 @@ parser.add_argument(
     help="The GCP region. This is used to inject a variable value into the notebook before running.",
     required=True,
 )
+parser.add_argument(
+    "--variable_service_account",
+    type=str,
+    help="A service account. This is used to inject a variable value into the notebook before running. This is not the account that will run the notebook.",
+    required=True,
+)
+parser.add_argument(
+    "--variable_vpc_network",
+    type=str,
+    help="The full VPC network name. See https://cloud.google.com/compute/docs/networks-and-firewalls#networks. Format is projects/{project}/global/networks/{network}, where {project} is a project number, as in '12345', and {network} is network name. See <https://cloud.google.com/compute/docs/reference/rest/v1/networks/insert> for details. This is used to inject a variable value into the notebook before running.",
+    required=False,
+)
 parser.add_argument(
     "--staging_bucket",
     type=str,
@@ -108,9 +120,11 @@ execute_changed_notebooks_helper.process_and_execute_notebooks(
     container_uri=args.container_uri,
     staging_bucket=args.staging_bucket,
     artifacts_bucket=args.artifacts_bucket,
-    variable_project_id=args.variable_project_id,
-    variable_region=args.variable_region,
-    private_pool_id=args.private_pool_id,
     should_parallelize=args.should_parallelize,
     timeout=args.timeout,
+    variable_project_id=args.variable_project_id,
+    variable_region=args.variable_region,
+    variable_service_account=args.variable_service_account,
+    variable_vpc_network=args.variable_vpc_network,
+    private_pool_id=args.private_pool_id,
 )

.cloud-build/execute_changed_notebooks_helper.py

@@ -17,6 +17,7 @@ import concurrent
 import dataclasses
 import datetime
 import functools
+import git
 import operator
 import os
 import pathlib
@@ -67,11 +68,20 @@ class NotebookExecutionResult:
     build_id: str
     error_message: Optional[str]
 
+    @property
+    def output_uri_web(self) -> Optional[str]:
+        if self.output_uri.startswith("gs://"):
+            return f"https://storage.googleapis.com/{self.output_uri[5:]}"
+        else:
+            return None
+
 
 def _process_notebook(
     notebook_path: str,
     variable_project_id: str,
     variable_region: str,
+    variable_service_account: str,
+    variable_vpc_network: Optional[str],
 ):
     # Read notebook
     with open(notebook_path) as f:
@@ -83,6 +93,8 @@ def _process_notebook(
         replacement_map={
             "PROJECT_ID": variable_project_id,
             "REGION": variable_region,
+            "SERVICE_ACCOUNT": variable_service_account,
+            "VPC_NETWORK": variable_vpc_network,
         },
     )
 
@@ -117,8 +129,10 @@ def process_and_execute_notebook(
     artifacts_bucket: str,
     variable_project_id: str,
     variable_region: str,
+    variable_service_account: str,
+    variable_vpc_network: Optional[str],
     private_pool_id: Optional[str],
-    deadline: datetime,
+    deadline: datetime.datetime,
     notebook: str,
     should_get_tail_logs: bool = False,
 ) -> NotebookExecutionResult:
@@ -126,6 +140,13 @@ def process_and_execute_notebook(
 
     print(f"Running notebook: {notebook}")
 
+    # Handle empty strings
+    if not variable_vpc_network:
+        variable_vpc_network = None
+
+    if not private_pool_id:
+        private_pool_id = None
+
     # Create paths
     notebook_output_uri = "/".join([artifacts_bucket, pathlib.Path(notebook).name])
 
@@ -151,6 +172,8 @@ def process_and_execute_notebook(
             notebook_path=notebook,
             variable_project_id=variable_project_id,
             variable_region=variable_region,
+            variable_service_account=variable_service_account,
+            variable_vpc_network=variable_vpc_network,
         )
 
         # Upload the pre-processed code to a GCS bucket
@@ -232,20 +255,40 @@ def get_changed_notebooks(
 
     # Find notebooks
     notebooks = []
+
+    # Instantiate GitPython objects
+    repo = git.Repo(os.getcwd())
+    index = repo.index
+
     if base_branch:
-        print(f"Looking for notebooks that changed from branch: {base_branch}")
-        notebooks = subprocess.check_output(
-            ["git", "diff", "--name-only", f"origin/{base_branch}..."] + test_paths
-        )
+        # Get the point at which this branch branches off from main
+        branching_commits = repo.merge_base("HEAD", f"origin/{base_branch}")
+
+        if len(branching_commits) > 0:
+            branching_commit = branching_commits[0]
+            print(f"Looking for notebooks that changed from branch: {branching_commit}")
+
+            notebooks = [
+                diff.b_path
+                for diff in index.diff(branching_commit, paths=test_paths)
+                if diff.b_path is not None
+            ]
+        else:
+            notebooks = []
     else:
         print(f"Looking for all notebooks.")
-        notebooks = subprocess.check_output(["git", "ls-files"] + test_paths)
+        notebooks_str = subprocess.check_output(["git", "ls-files"] + test_paths)
+        notebooks = notebooks_str.decode("utf-8").split("\n")
 
-    notebooks = notebooks.decode("utf-8").split("\n")
     notebooks = [notebook for notebook in notebooks if notebook.endswith(".ipynb")]
     notebooks = [notebook for notebook in notebooks if len(notebook) > 0]
     notebooks = [notebook for notebook in notebooks if pathlib.Path(notebook).exists()]
 
+    if len(notebooks) > 0:
+        print(f"Found {len(notebooks)} notebooks:")
+        for notebook in notebooks:
+            print(f"\t{notebook}")
+
     return notebooks
 
 
@@ -254,11 +297,13 @@ def process_and_execute_notebooks(
     container_uri: str,
     staging_bucket: str,
     artifacts_bucket: str,
-    variable_project_id: str,
-    variable_region: str,
-    private_pool_id: Optional[str],
     should_parallelize: bool,
     timeout: int,
+    variable_project_id: str,
+    variable_region: str,
+    variable_service_account: str,
+    variable_vpc_network: Optional[str] = None,
+    private_pool_id: Optional[str] = None,
 ):
     """
     Run the notebooks that exist under the folders defined in the test_paths_file.
@@ -315,6 +360,8 @@ def process_and_execute_notebooks(
                             artifacts_bucket,
                             variable_project_id,
                             variable_region,
+                            variable_service_account,
+                            variable_vpc_network,
                             private_pool_id,
                             deadline,
                         ),
@@ -329,6 +376,8 @@ def process_and_execute_notebooks(
                     artifacts_bucket=artifacts_bucket,
                     variable_project_id=variable_project_id,
                     variable_region=variable_region,
+                    variable_service_account=variable_service_account,
+                    variable_vpc_network=variable_vpc_network,
                     private_pool_id=private_pool_id,
                     deadline=deadline,
                     notebook=notebook,
@@ -354,10 +403,18 @@ def process_and_execute_notebooks(
                         format_timedelta(result.duration),
                         result.log_url,
                         result.output_uri,
+                        result.output_uri_web,
                     ]
                     for result in results_sorted
                 ],
-                headers=["build_tag", "status", "duration", "log_url", "output_url"],
+                headers=[
+                    "build_tag",
+                    "status",
+                    "duration",
+                    "log_url",
+                    "output_uri",
+                    "output_uri_web",
+                ],
             )
         )
 
@@ -385,6 +442,8 @@ def process_and_execute_notebooks(
             notebook_path=notebook,
             variable_project_id=variable_project_id,
             variable_region=variable_region,
+            variable_service_account=variable_service_account,
+            variable_vpc_network=variable_vpc_network,
         )
 
         execute_notebook_helper.execute_notebook(

.cloud-build/execute_notebook_helper.py

@@ -26,6 +26,9 @@ from utils import util
 
 # This script is used to execute a notebook and write out the output notebook.
 
+# This is used to force papermill to use this kernel to run the notebook instead of any defined inside the notebook itself
+DEFAULT_KERNEL_NAME = "python3"
+
 
 def execute_notebook(
     notebook_source: str,
@@ -50,6 +53,17 @@ def execute_notebook(
 
     execution_exception = None
 
+    print("\n=== DOWNLOAD EXECUTED NOTEBOOK ===\n")
+    print(f"Please debug the executed notebook by downloading the executed notebook:")
+
+    print("Option 1. Using gsutil. Run the following command in your terminal.")
+    print(f'\tgsutil cp "{output_file_or_uri}" .')
+
+    print("Option 2. Using this link.")
+    print(f"\thttps://storage.googleapis.com/{output_file_or_uri[5:]}")
+
+    print("\n======\n")
+
     # Execute notebook
     try:
         # Execute notebook
@@ -58,6 +72,7 @@ def execute_notebook(
             output_path=notebook_source,
             progress_bar=should_log_output,
             request_save_on_cell_execute=should_log_output,
+            kernel_name=DEFAULT_KERNEL_NAME,
             log_output=should_log_output,
             stdout_file=sys.stdout if should_log_output else None,
             stderr_file=sys.stderr if should_log_output else None,
@@ -71,10 +86,6 @@ def execute_notebook(
             util.upload_file(notebook_source, remote_file_path=output_file_or_uri)
 
             print("\n=== EXECUTION FINISHED ===\n")
-            print(
-                f"Please debug the executed notebook by downloading: {output_file_or_uri}"
-            )
-            print("\n======\n")
         else:
             # Create directories if they don't exist
             if not os.path.exists(os.path.dirname(output_file_or_uri)):

.cloud-build/notebook-execution-test-cloudbuild-single.yaml

@@ -4,25 +4,35 @@ steps:
     entrypoint: /bin/sh
     args:
     - -c
-    - 'gcloud config list'
+    - 'gcloud config list --quiet'
   # Check the Python version
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'python3 .cloud-build/CheckPythonVersion.py'
+    - python3 .cloud-build/CheckPythonVersion.py -q
+  # Create a virtual environment
+  - name: ${_PYTHON_IMAGE}
+    entrypoint: /bin/sh
+    args:
+    - -c
+    - python3 -m venv workspace/env
   # Install Python dependencies
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'python3 -m pip install -U pip && python3 -m pip install -U --user -r .cloud-build/requirements.txt'
+    - . workspace/env/bin/activate &&
+      python3 -m pip -q install -U pip &&
+      python3 -m pip -q install -U -r .cloud-build/requirements.txt
   # Install Python dependencies and run testing script
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'python3 -m pip install -U pip && python3 -m pip freeze && python3 .cloud-build/execute_notebook_cli.py --notebook_source "${_NOTEBOOK_GCS_URI}" --output_file_or_uri "${_NOTEBOOK_OUTPUT_GCS_URI}"'
+    - |
+      . workspace/env/bin/activate &&
+      python3 .cloud-build/execute_notebook_cli.py --notebook_source "${_NOTEBOOK_GCS_URI}" --output_file_or_uri "${_NOTEBOOK_OUTPUT_GCS_URI}"
     env:
     - 'IS_TESTING=1'
-timeout: 86400s
+timeout: 86400s

.cloud-build/notebook-execution-test-cloudbuild.yaml

@@ -4,35 +4,42 @@ steps:
     entrypoint: /bin/sh
     args:
     - -c
-    - 'gcloud config list'
+    - gcloud config list --quiet
   # Check the Python version
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'python3 .cloud-build/CheckPythonVersion.py'
-  # Fetch base branch if required
+    - python3 .cloud-build/CheckPythonVersion.py -q
+  # Fetch full repo for diff purposes
+  - name: gcr.io/cloud-builders/git
+    args: [fetch, --unshallow, --quiet]
+  # Create a virtual environment
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'if [ -n "${_BASE_BRANCH}" ]; then git fetch origin "${_BASE_BRANCH}":refs/remotes/origin/"${_BASE_BRANCH}"; else echo "Skipping fetch."; fi'
+    - python3 -m venv workspace/env
   # Install Python dependencies
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'python3 -m pip install -U pip && python3 -m pip install -U --user -r .cloud-build/requirements.txt'
+    - . workspace/env/bin/activate &&
+      python3 -m pip -q install -U pip &&
+      python3 -m pip -q install -U -r .cloud-build/requirements.txt
   # Install Python dependencies and run testing script
   # TODO: Only pass in private_pool_id if it is set
   - name: ${_PYTHON_IMAGE}
     entrypoint: /bin/sh
     args:
     - -c
-    - 'python3 -m pip install -U pip && python3 -m pip freeze && python3 .cloud-build/execute_changed_notebooks_cli.py --test_paths_file "${_TEST_PATHS_FILE}" --base_branch "${_FORCED_BASE_BRANCH}" --container_uri ${_PYTHON_IMAGE} --staging_bucket ${_GCS_STAGING_BUCKET}  --artifacts_bucket ${_GCS_STAGING_BUCKET}/executed_notebooks/PR_${_PR_NUMBER}/BUILD_${BUILD_ID} --variable_project_id ${PROJECT_ID} --variable_region ${_GCP_REGION} `if [ ! -z "${_PRIVATE_POOL_NAME}" ]; then echo "--private_pool_id ${_PRIVATE_POOL_NAME}"; fi`'
+    - |
+      . workspace/env/bin/activate &&
+      python3 .cloud-build/execute_changed_notebooks_cli.py --test_paths_file "${_TEST_PATHS_FILE}" --base_branch "${_FORCED_BASE_BRANCH}" --container_uri ${_PYTHON_IMAGE} --staging_bucket ${_GCS_STAGING_BUCKET}  --artifacts_bucket ${_GCS_STAGING_BUCKET}/executed_notebooks/PR_${_PR_NUMBER}/BUILD_${BUILD_ID} --variable_project_id ${PROJECT_ID} --variable_region ${_GCP_REGION} --variable_service_account ${_GCP_SERVICE_ACCOUNT} --variable_vpc_network "${_GPC_VPC_NETWORK_NAME}" `if [ ! -z "${_PRIVATE_POOL_NAME}" ]; then echo "--private_pool_id ${_PRIVATE_POOL_NAME}"; fi`
     env:
     - 'IS_TESTING=1'
 timeout: 86400s
 options:
   pool:
-    name: ${_PRIVATE_POOL_NAME}
+    name: ${_PRIVATE_POOL_NAME}

.cloud-build/requirements.txt

@@ -10,3 +10,4 @@ google-cloud-aiplatform
 google-cloud-storage
 google-cloud-build
 ratemate
+GitPython

.cloud-build/test_notebook_vm.txt

@@ -1,5 +1,5 @@
 notebooks/official/vizier/gapic-vizier-multi-objective-optimization.ipynb
 notebooks/official/pipelines/lightweight_functions_component_io_kfp.ipynb
-notebooks/official/matching_engine/intro-swivel.ipynb
 notebooks/official/ml_metadata/sdk-metric-parameter-tracking-for-locally-trained-models.ipynb
 notebooks/official/pipelines/metrics_viz_run_compare_kfp.ipynb
+notebooks/official/matching_engine/sdk_matching_engine_for_indexing.ipynb

.cloud-build/test_single_notebook_vm.txt

@@ -0,0 +1 @@
+notebooks/official/pipelines/metrics_viz_run_compare_kfp.ipynb

.cloud-build/utils/UpdateNotebookVariables.py

@@ -35,8 +35,8 @@ Variables in conditionals can also be replaced:
 
 def get_updated_value(content: str, variable_name: str, variable_value: str) -> str:
     return re.sub(
-        rf"({variable_name}.*?=.*?[\",\'])\[.+?\]([\",\'].*?)",
-        rf"\1{variable_value}\2",
+        rf"({variable_name}.*? = .*?[\",\'])\[.+?\]([\",\'].*?)",
+        rf"\g<1>{variable_value}\g<2>",
         content,
         flags=re.M,
     )
@@ -79,3 +79,26 @@ def test_region():
         variable_value="us-central1",
     )
     assert new_content == 'REGION = "us-central1"  # @param {type:"string"}'
+
+
+def test_region_equal_equals_ignore():
+    # Tests that == is ignored
+    new_content = get_updated_value(
+        content='REGION == "[your-region]"  # @param {type:"string"}',
+        variable_name="REGION",
+        variable_value="us-central1",
+    )
+    assert new_content == 'REGION == "[your-region]"  # @param {type:"string"}'
+
+
+def test_service_account():
+    # Tests that == is ignored
+    new_content = get_updated_value(
+        content='SERVICE_ACCOUNT = "[your-service-account]"  # @param {type:"string"}',
+        variable_name="SERVICE_ACCOUNT",
+        variable_value="12345-compute@developer.gserviceaccount.com",
+    )
+    assert (
+        new_content
+        == 'SERVICE_ACCOUNT = "12345-compute@developer.gserviceaccount.com"  # @param {type:"string"}'
+    )

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,4 +1,11 @@
-If you are opening a PR for `Official Notebooks` under the [notebooks/official](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/official) folder, follow this mandatory checklist:
+**REQUIRED:** Add a summary of your PR here, typically including why the change is needed and what was changed. Include any design alternatives for discussion purposes.
+
+<br>
+--- YOUR PR SUMMARY GOES HERE ---
+<br><br><br>
+
+**REQUIRED:** Fill out the below checklists or remove if irrelevant
+1. If you are opening a PR for `Official Notebooks` under the [notebooks/official](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/official) folder, follow this mandatory checklist:
 - [ ] Use the [notebook template](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/notebook_template.ipynb) as a starting point.
 - [ ] Follow the style and grammar rules outlined in the above notebook template.
 - [ ] Verify the notebook runs successfully in Colab since the automated tests cannot guarantee this even when it passes.
@@ -7,12 +14,15 @@ If you are opening a PR for `Official Notebooks` under the [notebooks/official](
 - [ ] This notebook has been added to the [CODEOWNERS](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/CODEOWNERS) file under the `Official Notebooks` section, pointing to the author or the author's team.
 - [ ] The Jupyter notebook cleans up any artifacts it has created (datasets, ML models, endpoints, etc) so as not to eat up unnecessary resources.
 
+<br>
 
-If you are opening a PR for `Community Notebooks` under the [notebooks/community](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community) folder:
+2. If you are opening a PR for `Community Notebooks` under the [notebooks/community](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community) folder:
 - [ ] This notebook has been added to the [CODEOWNERS](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/CODEOWNERS) file under the `Community Notebooks` section, pointing to the author or the author's team.
 - [ ] Passes all the required formatting and linting checks. You can locally test with these [instructions](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/CONTRIBUTING.md#code-quality-checks).
 
-If you are opening a PR for `Community Content` under the [community-content](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/community-content) folder:
+<br>
+
+3. If you are opening a PR for `Community Content` under the [community-content](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/community-content) folder:
 - [ ] Make sure your main `Content Directory Name` is descriptive, informative, and includes some of the key products and attributes of your content, so that it is differentiable from other content
 - [ ] The main content directory has been added to the [CODEOWNERS](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/community-content/CODEOWNERS) file under the `Community Content` section, pointing to the author or the author's team.
 - [ ] Passes all the required formatting and linting checks. You can locally test with these [instructions](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/CONTRIBUTING.md#code-quality-checks).

.github/workflows/ci.yaml

@@ -7,7 +7,9 @@ jobs:
     runs-on: ubuntu-latest
     steps:
     - name: Set up Python
-      uses: actions/setup-python@v3
+      uses: actions/setup-python@v4
+      with:
+        python-version: '3.x'      
     - name: Fetch pull request branch
       uses: actions/checkout@v3
       with:

.github/workflows/linter/requirements.txt

@@ -2,9 +2,9 @@ git+https://github.com/tensorflow/docs
 ipython
 jupyter
 nbconvert
-black==22.3.0
-pyupgrade==2.31.1
+black==22.6.0
+pyupgrade==2.34.0
 isort==5.10.1
 flake8==4.0.1
-nbqa==1.3.1
-protobuf ==3.19.0
+nbqa==1.4.0
+

.github/workflows/linter/run_linter.sh

@@ -68,7 +68,7 @@ if [ ${#notebooks[@]} -gt 0 ]; then
 
             if [ "$is_test" = true ]; then
                 echo "Running nbfmt..."
-                python3 -m tensorflow_docs.tools.nbfmt --remove_outputs --test "$notebook"
+                python3 -m tensorflow_docs.tools.nbfmt --test "$notebook"
                 NBFMT_RTN=$?
                 # echo "Running black..."
                 # python3 -m nbqa black "$notebook" --check
@@ -93,7 +93,7 @@ if [ ${#notebooks[@]} -gt 0 ]; then
                 python3 -m nbqa isort "$notebook"
                 ISORT_RTN=$?
                 echo "Running nbfmt..."
-                python3 -m tensorflow_docs.tools.nbfmt --remove_outputs "$notebook"
+                python3 -m tensorflow_docs.tools.nbfmt "$notebook"
                 NBFMT_RTN=$?
                 echo "Running flake8..."
                 python3 -m nbqa flake8 "$notebook" --show-source --extend-ignore=W391,E501,F821,E402,F404,W503,E203,E722,W293,W291

CONTRIBUTING.md

@@ -48,8 +48,8 @@ then you will need to manually address them before submitting your PR.
 nbqa black "$notebook"
 nbqa pyupgrade "$notebook"
 nbqa isort "$notebook"
-python3 -m tensorflow_docs.tools.nbfmt --remove_outputs "$notebook"
 nbqa flake8 "$notebook" --extend-ignore=W391,E501,F821,E402,F404,W503,E203,E722,W293,W291
+python3 -m tensorflow_docs.tools.nbfmt --remove_outputs "$notebook"
 ```
 
 ## Code Reviews

community-content/CODEOWNERS

@@ -1,6 +1,8 @@
 *  @vertex-ai-samples-contributors  @GoogleCloudPlatform/cloudml-samples-owners
 /tf_agents_bandits_movie_recommendation_with_kfp_and_vertex_sdk	@yinghsienwu
+/pytorch_pre_built_images_deployment @googleapis/vertex-prediction-team
 /pytorch_text_classification_using_vertex_sdk_and_gcloud  @RajeshThallam
 /pytorch_text_classification_using_vertex_sdk_and_gcloud		@RajeshThallam @ultrons
 /sklearn_text_classification_from_script_using_vertex_sdk       @maxhardt
 /pluto_on_workbench @wkharold
+/cpr-examples @samthrasher

community-content/cpr-examples/timm_serving/.dockerignore

@@ -0,0 +1,5 @@
+testdata/*
+build.py
+test.py
+state_dict.pth
+config.json

community-content/cpr-examples/timm_serving/.gitignore

@@ -0,0 +1,5 @@
+cpr_model_server.py
+entrypoint.py
+state_dict.pth
+config.json
+**/__pycache__

community-content/cpr-examples/timm_serving/README.md

@@ -0,0 +1,93 @@
+# CPR Example: PyTorch Image Models (timm)
+
+## About CPR
+
+CPR ([custom prediction routines](https://github.com/googleapis/python-aiplatform/blob/custom-prediction-routine/google/cloud/aiplatform/prediction/README.md)) is a framework designed by Google Cloud developers to make it easier to combine machine learning models with custom preprocessing and postprocessing logic in a real-time serving application. 
+
+## Using this example
+
+This code is a self-contained example of a custom model server project built using CPR.
+
+As is, you can use it to serve the ViT-Small image classification model from Ross Wightman's [`timm`](https://github.com/rwightman/pytorch-image-models) library of image model implementations in PyTorch. Both CPU and GPU are supported.
+
+You can also consider using the code here as a template for your own CPR project if you want to use a different model from `timm`, a different PyTorch model, or an entirely different framework.
+
+### Requirements
+
+In order to use this example, you'll need Docker and Python 3 installed on your system.
+
+To get started, first create a virtual environment in an empty directory:
+```sh
+mkdir cpr-example
+python3 -m venv cpr-example
+cd cpr-example && source bin/activate
+``` 
+
+Then, clone the [vertex-ai-samples repo](https://github.com/GoogleCloudPlatform/vertex-ai-samples) in that directory:
+```sh
+git clone https://github.com/GoogleCloudPlatform/vertex-ai-samples.git
+cd vertex-ai-samples/community-content/cpr-examples/timm_serving
+```
+
+Finally, install the Python modules required to build and run the model server:
+```sh
+pip install -r requirements.txt
+```
+
+### Predictor
+
+The `TimmPredictor` class in `timm_serving/predictor.py` implements most of the important logic for the server.
+
+- `load(artifacts_dir)`: The predictor's `load` method is called when the server starts up in order to set up the predictor, usually by loading model weights and any artifacts needed for preprocessing and postprocessing. In this example, we initialize the saved model from the `state_dict.pth` file located inside the `artifacts_dir` folder and create the preprocessing transform from the model config.
+
+- `preprocess`, `predict`, `postprocess`: These methods are applied in sequence to the deserialized JSON data from each request. 
+    - `preprocess` decodes images from base64 and apply cropping, scaling and normalizing transforms. 
+    - `predict` runs the ViT-Small model on the preprocessed images and returns class scores.
+    - `postprocess` finds the top five classes and packs the class names, probabilities, and indices in a serializable result.
+
+### Building the container
+
+To build the model server locally, run the build command:
+```sh
+python build.py build
+```
+
+You can edit configuration values such as the model server's base image, the name and tag assigned to the image, and the path where model weights are stored locally.
+
+When you run the build command, model weights are downloaded and the model server container is built.
+
+### Running local tests
+
+`test.py` contains a suite of unit tests for the predictor as well as end-to-end tests for the model server. 
+
+To run the tests:
+```sh
+python test.py
+```
+
+All of the test images are public domain.
+- [Cat](https://commons.wikimedia.org/wiki/File:Stray_cat_on_wall.jpg)
+- [Airplane](https://commons.wikimedia.org/wiki/File:Airplanes_jets.jpg)
+- The infamous [mandrill](https://commons.wikimedia.org/wiki/File:Wikipedia-sipi-image-db-mandrill-4.2.03.png)
+
+### Deploying to Vertex AI
+
+Before uploading or deploying the container, you'll need to modify `config.py` to set appropriate values for: 
+- `project_id`: Your GCP project id.
+- `region`: Region where the model will be uploaded and deployed.
+- `repository`: [Artifact Registry repository](https://cloud.google.com/artifact-registry/docs/repositories/create-repos) in your project where the container image will be uploaded. 
+- `artifacts_gcs_dir`: Folder in a [Google Cloud Storage bucket](https://cloud.google.com/storage/docs/creating-buckets) where the model weights will be uploaded.
+
+Once this is done, first upload the model:
+```sh
+python build.py upload
+```
+
+Then deploy it:
+```sh
+python build.py deploy
+```
+
+If you run the deploy command again, it will create a new endpoint. If you want to undeploy the model, you can do so using the Vertex AI dashboard on the Google Cloud console, or use `gcloud ai endpoints undeploy` from the command line.
+
+After deploying successfully, you can run `python build.py probe` to send a sample request to the deployed model. 

community-content/cpr-examples/timm_serving/build.py

@@ -0,0 +1,117 @@
+# Copyright 2022 Google LLC
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+#     https://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Build the model server container."""
+import json
+import logging
+import os
+import pathlib
+from typing import Sequence
+
+from absl import app
+from absl import logging
+from config import CPRConfig
+from google.cloud import aiplatform
+from google.cloud.aiplatform import prediction as cpr
+import smart_open
+import timm
+from timm_serving import predictor
+import torch
+
+
+def build_container(config: CPRConfig, tag: str) -> cpr.LocalModel:
+    """Build the model server container.
+
+    Args:
+      tag: Output image tag.
+
+    Returns:
+      LocalModel exposing the built model server.
+    """
+    return cpr.LocalModel.build_cpr_model(
+        src_dir=os.path.join(os.getcwd()),
+        output_image_uri=tag,
+        base_image=config.base_image,
+        predictor=predictor.TimmPredictor,
+        requirements_path=os.path.join(os.getcwd(), "requirements.txt"),
+    )
+
+
+def save_model_artifact(destination: str) -> None:
+    """Save a copy of the model state dict."""
+    model = timm.create_model(predictor.TimmPredictor.TIMM_MODEL_NAME, pretrained=True)
+    dest_file = os.path.join(destination, predictor.TimmPredictor.WEIGHTS_FILE)
+    with smart_open.open(dest_file, "wb") as f:
+        torch.save(model, f)
+    logging.info("Saved model to %s", dest_file)
+    logging.info("%s parameters", sum(p.numel() for p in model.parameters()))
+
+
+def upload_model(config: CPRConfig) -> aiplatform.Model:
+    """Tag and upload the model server."""
+    ar_tag = (
+        f"{config.region}-docker.pkg.dev/{config.project_id}"
+        f"/{config.repository}/{config.image}"
+    )
+    local_model = build_container(config, tag=ar_tag)
+    aiplatform.init(project=config.project_id, location=config.region)
+    local_model.push_image()
+    aip_model = aiplatform.Model.upload(
+        local_model=local_model,
+        display_name=predictor.TimmPredictor.TIMM_MODEL_NAME,
+        artifact_uri=config.artifact_gcs_dir,
+    )
+    config.model_name = aip_model.resource_name
+    config.save()
+    return aip_model
+
+
+def deploy_model(config: CPRConfig) -> aiplatform.Endpoint:
+    """Deploy the model server to a Vertex Prediction endpoint."""
+    aiplatform.init(project=config.project_id, location=config.region)
+    aip_model = aiplatform.Model(model_name=config.model_name)
+    endpoint = aip_model.deploy(machine_type=config.machine_type)
+    config.endpoint_name = endpoint.resource_name
+    config.save()
+    return endpoint
+
+
+def probe_prediction(config: CPRConfig, request_path: str) -> None:
+    """Send a sample prediction request to the Vertex Prediction endpoint."""
+    aiplatform.init(project=config.project_id, location=config.region)
+    aip_endpoint = aiplatform.Endpoint(endpoint_name=config.endpoint_name)
+    with open(request_path) as f:
+        logging.info(aip_endpoint.predict(**json.load(f)))
+
+
+def main(argv: Sequence[str]):
+    config = CPRConfig()
+    if pathlib.Path(config.config_file).exists():
+        config.load()
+
+    actions = set(argv[1:])
+    if "build" in actions:
+        build_container(config, config.image)
+        save_model_artifact(config.artifact_local_dir)
+    if "upload" in actions:
+        save_model_artifact(config.artifact_gcs_dir)
+        upload_model(config)
+    if "deploy" in actions:
+        deploy_model(config)
+    if "probe" in actions:
+        probe_prediction(config, request_path="sample_request.json")
+
+
+if __name__ == "__main__":
+    app.run(main)

community-content/cpr-examples/timm_serving/config.py

@@ -0,0 +1,76 @@
+# Copyright 2022 Google LLC
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+#     https://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import dataclasses
+import json
+
+
+@dataclasses.dataclass
+class CPRConfig(object):
+    """Configure the build process by editing the default values here.
+
+    config_file: File path used to save values in this config. (Some
+      values, such as the model name, are generated at build time and
+      depended on by future steps, so saving it allows this script to
+      deploy the model without re-uploading it, for example.)
+
+    base_image: Base Docker image on top of which the model server will
+      be built. By default, a Debian-based Python 3 image without GPU
+      support will be used.
+
+    image: Name and tag assigned to the built model server image.
+
+    artifact_local_dir: Local directory where a copy of the pretrained model weights
+      will be saved.
+
+    region: Google Cloud Region where the model will be uploaded during the
+      build process.
+
+    project_id: Google Cloud project ID.
+
+    repository: Name of the Artifact Registry repository where the container
+      will be uploaded.
+
+    artifact_gcs_dir: Location on GCS where a copy of the pretrained model
+      weights will be uploaded.
+
+    model_name: Full resource path of the uploaded model. This is a write-only
+      field, the value is generated by Vertex AI when the model is uploaded.
+
+    endpoint_name: Full resource path of the created endpoint. This is a
+      write-only field, the value is generated by Vertex AI when the model is
+      deployed to an endpoint.
+
+    machine_type: Machine type to use when deploying the model.
+    """
+
+    config_file: str = "config.json"
+    base_image: str = "python:3.10-bullseye"
+    image: str = "timm_predictor:latest"
+    artifact_local_dir: str = ""
+    region: str = "us-central1"
+    project_id: str = "samthrasher-experimental"
+    repository: str = "cpr-images"
+    artifact_gcs_dir: str = "gs://samthrasher-cpr-example/timm-vit224/"
+    model_name: str = ""
+    endpoint_name: str = ""
+    machine_type: str = "n1-standard-2"
+
+    def save(self):
+        with open(self.config_file, "w") as f:
+            json.dump(dataclasses.asdict(self), f, indent=2)
+
+    def load(self):
+        with open(self.config_file) as f:
+            self.__init__(**json.load(f))

community-content/cpr-examples/timm_serving/requirements.txt

@@ -0,0 +1,8 @@
+absl-py==1.1.0
+fastapi==0.75.2
+uvicorn==0.18.2
+timm==0.5.4
+smart_open==6.0.0
+
+google-cloud-storage>=1.26.0,<2.0.0dev
+google-cloud-aiplatform[prediction] @ git+https://github.com/googleapis/python-aiplatform.git@custom-prediction-routine

community-content/cpr-examples/timm_serving/test.py

@@ -0,0 +1,249 @@
+"""Test the timm_serving predictor."""
+import base64
+import json
+import logging
+import os
+import pickle
+from typing import List, Dict
+
+from absl import flags
+from absl import logging
+from absl.testing import absltest
+from config import CPRConfig
+import fastapi
+from google.cloud import aiplatform
+from google.cloud.aiplatform import prediction as cpr
+import PIL
+from timm_serving import predictor
+import torch
+
+VIT_SMALL_PARAMS = 22878952
+
+
+def b64_encode_file(path: str) -> str:
+    """Encode a file's contents as base64.
+
+    Args:
+      path: Path to the file.
+
+    Returns:
+      Base64-encoded contents of the file.
+    """
+    with open(path, "rb") as f:
+        return str(base64.b64encode(f.read()), encoding="utf-8")
+
+
+def make_instance_dict(
+    image_paths: List[str], base64_encodings: List[str]
+) -> Dict[str, List[str]]:
+    """Generate a dictionary similar to a parsed prediction server request.
+
+    Args:
+      image_paths: Paths to image files to include.
+      base64_encodings: Pre-encoded base64 strings.
+
+    Returns:
+      Dictionary of instances in the format accepted by the preprocessor.
+    """
+    instances = [s for s in base64_encodings]
+    for path in image_paths:
+        instances.append(b64_encode_file(path))
+    return {"instances": instances}
+
+
+def count_parameters(model: torch.nn.Module):
+    """Count the parameters in a Pytorch model.
+
+    Args:
+      model: Pytorch model (nn.Module).
+
+    Returns:
+      Number of parameters in the model.
+
+    """
+    return sum(p.numel() for p in model.parameters())
+
+
+class PredictorUnitTests(absltest.TestCase):
+    """Unit tests for timm_serving.predictor."""
+
+    def setUp(self):
+        super().setUp()
+        self.config = CPRConfig()
+        self.config.load()
+        self.predictor = predictor.TimmPredictor()
+
+    def test_load_from_saved_state_dict_ok(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        self.assertEqual(count_parameters(self.predictor._model), VIT_SMALL_PARAMS)
+
+    def test_load_bad_path(self):
+        with self.assertRaises(FileNotFoundError):
+            self.predictor.load("testdata/")
+        with self.assertRaisesRegex(ValueError, "not a directory"):
+            self.predictor.load("blah")
+
+    def test_load_bad_data(self):
+        with self.assertRaises(pickle.UnpicklingError):
+            self.predictor.load("testdata/bad_model_1")
+        with self.assertRaisesRegex(RuntimeError, "Invalid magic number"):
+            self.predictor.load("testdata/bad_model_2")
+
+    def test_preprocess_ok(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        instance_dict = make_instance_dict(
+            base64_encodings=[],
+            image_paths=[
+                "testdata/airplane.jpg",
+                "testdata/mandrill.tiff",
+                "testdata/mandrill.tiff",
+                "testdata/cat_alpha.png",
+            ],
+        )
+        result = self.predictor.preprocess(instance_dict)
+        self.assertEqual(result.size(), torch.Size([4, 3, 224, 224]))
+        self.assertEqual(result.dtype, torch.float32)
+
+    def test_preprocess_no_instances(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        with self.assertRaises(fastapi.HTTPException) as ctx:
+            self.predictor.preprocess({})
+        self.assertEqual(ctx.exception.status_code, 400)
+        self.assertRegex(ctx.exception.detail, 'must contain "instances"')
+
+    def test_preprocess_wrong_shape_instances(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        instance_dict = {"instances": [[b64_encode_file("testdata/mandrill.tiff")]]}
+        with self.assertRaises(fastapi.HTTPException) as ctx:
+            self.predictor.preprocess(instance_dict)
+        self.assertEqual(ctx.exception.status_code, 400)
+        self.assertRegex(ctx.exception.detail, "not 'list'")
+
+    def test_preprocess_bad_base64(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        instance_dict = make_instance_dict(base64_encodings=["!@#$"], image_paths=[])
+        with self.assertRaises(fastapi.HTTPException) as ctx:
+            self.predictor.preprocess(instance_dict)
+        self.assertEqual(ctx.exception.status_code, 400)
+        self.assertRegex(ctx.exception.detail, "[Bb]ase64")
+
+    def test_preprocess_not_image_data(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        instance_dict = make_instance_dict(
+            base64_encodings=[], image_paths=["testdata/bad.jpg"]
+        )
+        with self.assertRaises(fastapi.HTTPException) as ctx:
+            self.predictor.preprocess(instance_dict)
+        self.assertEqual(ctx.exception.status_code, 400)
+        self.assertRegex(ctx.exception.detail, "image file")
+
+    def test_predict_ok(self):
+        self.predictor.load(self.config.artifact_local_dir)
+        inputs = torch.zeros(size=[2, 3, 224, 224], dtype=torch.float32)
+        if torch.cuda.device_count() > 0:
+            inputs = inputs.cuda()
+        result = self.predictor.predict(inputs)
+        self.assertEqual(result.size(), torch.Size([2, 1000]))
+        self.assertEqual(result.dtype, torch.float32)
+
+    def test_postprocess_ok(self):
+        class_probs = torch.zeros(size=[2, 1000])
+        class_probs[0, 0] = 1
+        class_probs[1, 123] = 1
+        result = self.predictor.postprocess(class_probs)
+        predictions = result["predictions"]
+        self.assertLen(predictions[0]["class_names"], 5)
+        self.assertLen(predictions[0]["indices"], 5)
+        self.assertLen(predictions[0]["probabilities"], 5)
+        self.assertLen(predictions[1]["class_names"], 5)
+        self.assertLen(predictions[1]["indices"], 5)
+        self.assertLen(predictions[1]["probabilities"], 5)
+        self.assertContainsSubsequence(predictions[0]["class_names"][0], "tench")
+        self.assertContainsSubsequence(
+            predictions[1]["class_names"][0], "spiny lobster"
+        )
+
+
+class ServerEndToEndTests(absltest.TestCase):
+    """End-to-end tests for the model server, using LocalEndpoint."""
+
+    def setUp(self):
+        super().setUp()
+        self.config = CPRConfig()
+        self.config.load()
+        self.local_model = cpr.LocalModel(
+            serving_container_spec=aiplatform.gapic.ModelContainerSpec(
+                image_uri=self.config.image
+            )
+        )
+
+        self.local_endpoint = self.local_model.deploy_to_local_endpoint(
+            artifact_uri=self.config.artifact_local_dir or os.getcwd()
+        )
+        self.local_endpoint.serve()
+
+    def tearDown(self):
+        self.local_endpoint.stop()
+        super().tearDown()
+
+    def test_e2e_healthcheck_ok(self):
+        health_check_response = self.local_endpoint.run_health_check()
+        self.assertEqual(health_check_response.status_code, 200)
+        self.assertEqual(health_check_response.content, b"{}")
+
+    def test_e2e_predict_ok(self):
+        predict_request = json.dumps(
+            make_instance_dict(
+                base64_encodings=[],
+                image_paths=[
+                    "testdata/mandrill.tiff",
+                ],
+            )
+        )
+        response = self.local_endpoint.predict(
+            request=predict_request, headers={"Content-Type": "application/json"}
+        )
+        logging.info(response.content)
+        self.assertEqual(response.status_code, 200)
+        predictions = response.json()["predictions"]
+        self.assertContainsSubsequence(predictions[0]["class_names"][0], "baboon")
+
+    def test_e2e_predict_bad_json_returns_400(self):
+        predict_request = "blah"
+        response = self.local_endpoint.predict(
+            request=predict_request, headers={"Content-Type": "application/json"}
+        )
+        logging.info(response.content)
+        self.assertEqual(response.status_code, 400)
+
+    def test_e2e_predict_no_instances_returns_400(self):
+        predict_request = json.dumps({})
+        response = self.local_endpoint.predict(
+            request=predict_request, headers={"Content-Type": "application/json"}
+        )
+        logging.info(response.content)
+        self.assertEqual(response.status_code, 400)
+
+    def test_e2e_predict_bad_base64_returns_400(self):
+        predict_request = json.dumps(
+            make_instance_dict(base64_encodings=["blah"], image_paths=[])
+        )
+        response = self.local_endpoint.predict(
+            request=predict_request, headers={"Content-Type": "application/json"}
+        )
+        logging.info(response.content)
+        self.assertEqual(response.status_code, 400)
+
+    def test_e2e_predict_bad_image_returns_400(self):
+        predict_request = json.dumps(
+            make_instance_dict(base64_encodings=[], image_paths=["testdata/bad.jpg"])
+        )
+        response = self.local_endpoint.predict(
+            request=predict_request, headers={"Content-Type": "application/json"}
+        )
+        logging.info(response.content)
+        self.assertEqual(response.status_code, 400)
+
+
+if __name__ == "__main__":
+    absltest.main()

community-content/cpr-examples/timm_serving/testdata/bad.jpg

@@ -0,0 +1 @@
+some non-image data

community-content/cpr-examples/timm_serving/testdata/bad.txt

@@ -0,0 +1 @@
+some non-image data

community-content/cpr-examples/timm_serving/timm_serving/predictor.py

@@ -0,0 +1,178 @@
+# Copyright 2022 Google LLC
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+#     https://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Adapts a pretrained TIMM image classification model to the CPR framework.
+
+Documentation for the TIMM (Torch IMage Models) library is here:
+  https://rwightman.github.io/pytorch-image-models/
+
+Its source can also be found here:
+  https://github.com/rwightman/pytorch-image-models
+"""
+
+import base64
+import binascii
+import io
+import os
+from typing import Dict, List, Union
+
+from fastapi import HTTPException
+from google.cloud.aiplatform import prediction as cpr
+from pathlib import Path
+import PIL
+import smart_open
+import timm
+import torch
+import torch.nn.functional as F
+
+with open(Path(__file__).parent.absolute().joinpath("imagenet.txt")) as f:
+    IMAGENET_CLASSES = f.read().splitlines()
+
+
+class TimmPredictor(cpr.predictor.Predictor):
+    """Predictor class for image models based on TIMM."""
+
+    TIMM_MODEL_NAME = os.getenv("TIMM_MODEL_NAME", default="vit_small_patch32_224")
+    WEIGHTS_FILE = "state_dict.pth"
+    NUM_TOP_CLASSES_TO_RETURN = 5
+
+    def __init__(self):
+        self._cuda = torch.cuda.device_count() > 0
+
+    def load(self, artifacts_uri: str = ""):
+        """Initializes the model and preprocessing transforms.
+
+        Args:
+          artifacts_uri: Directory where state dict is stored. Can be a
+          GCS URI or local path.
+        """
+        if artifacts_uri:
+            artifact_path = os.path.join(artifacts_uri)
+            if not (os.path.isdir(artifact_path) or artifact_path.startswith("gs://")):
+                raise ValueError("Provided artifact_uri is not a directory.")
+        else:
+            artifact_path = os.getcwd()
+
+        artifact_path = os.path.join(artifact_path, self.WEIGHTS_FILE)
+        with smart_open.open(artifact_path, "rb") as f:
+            self._model = torch.load(f)
+
+        if self._cuda:
+            self._model.cuda()
+
+        config = timm.data.resolve_data_config(model=self.TIMM_MODEL_NAME, args=[])
+        self._transform = timm.data.create_transform(
+            is_training=False, use_prefetcher=False, **config
+        )
+
+    def preprocess(self, request_dict: Dict[str, List[str]]) -> torch.Tensor:
+        """Performs preprocessing.
+
+        By default, the server expects a request body consisting of a valid JSON
+        object. This will be parsed by the handler before it's evaluated by the
+        preprocess method.
+
+        Args:
+          request_dict: Parsed request body. We expect that the input consists of
+            a list of base64-encoded image files under the "instances" key. (Any
+            image format that PIL.image.open can handle is okay.)
+
+        Returns:
+          torch.Tensor containing the preprocessed images as a batch. If GPU is
+          available, the result tensor will be stored on GPU.
+        """
+
+        if "instances" not in request_dict:
+            raise HTTPException(
+                status_code=400,
+                detail='Request must contain "instances" as a top-level key.',
+            )
+
+        tensors = []
+
+        for (i, image) in enumerate(request_dict["instances"]):
+            # We use Base64 encoding to handle image data.
+            # This is probably the best we can do while still using JSON input.
+            # Overriding the input format requires building a custom Handler.
+            try:
+                image_bytes = base64.b64decode(image, validate=True)
+            except (binascii.Error, TypeError) as e:
+                raise HTTPException(
+                    status_code=400,
+                    detail=f"Base64 decoding of the input image at index {i} failed:"
+                    f" {str(e)}",
+                )
+
+            try:
+                pil_image = PIL.Image.open(io.BytesIO(image_bytes)).convert("RGB")
+            except PIL.UnidentifiedImageError:
+                raise HTTPException(
+                    status_code=400,
+                    detail=f"The input image at index {i} could not be identified as an"
+                    " image file.",
+                )
+
+            tensors.append(self._transform(pil_image))
+
+        with torch.inference_mode():
+            result = torch.stack(tensors)
+            if self._cuda:
+                result = result.cuda()
+        return result
+
+    def predict(self, instances: torch.Tensor) -> torch.Tensor:
+        """Performs prediction.
+
+        Args:
+          instances: torch.Tensor with type torch.float32 and shape
+            [?, 3, 224, 224], containing the pre-processed input images.
+
+        Returns:
+          Vector of scores with type torch.float32 and shape [?, 1000],
+            representing the model's estimate of the likelihood that the
+            input belongs to the Imagenet class with that index.
+        """
+        with torch.inference_mode():
+            class_scores = self._model(instances)
+        return class_scores
+
+    def postprocess(
+        self, class_scores: torch.Tensor
+    ) -> Dict[str, List[Dict[str, Union[str, int, float]]]]:
+        """Translate the model output into a classification result.
+
+        Args:
+          class_scores: torch.Tensor with type torch.float32 and shape
+            [?, 1000], containing the scores assigned to each class by
+            the model.
+
+        Returns:
+          Dictionary containing the list of classification results. Each
+            classification result contains the probabilities, class names, and
+            class indices of the classes with the top class scores as reported by
+            the model.
+        """
+        class_probs = F.softmax(class_scores, dim=1)
+        top_k = class_probs.topk(self.NUM_TOP_CLASSES_TO_RETURN)
+        top_k_values = top_k.values.numpy().tolist()
+        top_k_indices = top_k.indices.numpy().tolist()
+        predictions = [
+            dict(
+                probabilities=values,
+                indices=indices,
+                class_names=[IMAGENET_CLASSES[int(class_num)] for class_num in indices],
+            )
+            for (values, indices) in zip(top_k_values, top_k_indices)
+        ]
+        return {"predictions": predictions}

community-content/pytorch_image_classification_distributed_data_parallel_training_with_vertex_sdk/multi_node_ddp_gloo_vertex_training_with_custom_container.ipynb

@@ -1,474 +0,0 @@
-{
-  "cells": [
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "a6b56b1c7b76"
-      },
-      "outputs": [],
-      "source": [
-        "# Copyright 2021 Google LLC\n",
-        "#\n",
-        "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
-        "# you may not use this file except in compliance with the License.\n",
-        "# You may obtain a copy of the License at\n",
-        "#\n",
-        "#     https://www.apache.org/licenses/LICENSE-2.0\n",
-        "#\n",
-        "# Unless required by applicable law or agreed to in writing, software\n",
-        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
-        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
-        "# See the License for the specific language governing permissions and\n",
-        "# limitations under the License."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "c414a395a19b"
-      },
-      "source": [
-        "# PyTorch Image Classification Multi-Node Distributed Data Parallel Training on CPU using Vertex Training with Custom Container"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "b98238e32cf7"
-      },
-      "source": [
-        "<table align=\"left\">\n",
-        "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/master/community-content/pytorch_image_classification_distributed_data_parallel_training_with_vertex_sdk/multi_node_ddp_gloo_vertex_training_with_custom_container.ipynb\">\n",
-        "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
-        "      View on GitHub\n",
-        "    </a>\n",
-        "  </td>\n",
-        "</table>"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "03d216c7f7b1"
-      },
-      "source": [
-        "## Setup"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "c5ac73516218"
-      },
-      "outputs": [],
-      "source": [
-        "PROJECT_ID = \"YOUR PROJECT ID\"\n",
-        "BUCKET_NAME = \"gs://YOUR BUCKET NAME\"\n",
-        "REGION = \"YOUR REGION\"\n",
-        "SERVICE_ACCOUNT = \"YOUR SERVICE ACCOUNT\""
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "0b5ae674177e"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil ls -al $BUCKET_NAME"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "19a9b3bdd553"
-      },
-      "outputs": [],
-      "source": [
-        "content_name = \"pt-img-cls-multi-node-ddp-cust-cont\""
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "57bf6f8b4361"
-      },
-      "source": [
-        "## Local Training"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "e5d8a3443da0"
-      },
-      "outputs": [],
-      "source": [
-        "! ls trainer"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "07f79309472d"
-      },
-      "outputs": [],
-      "source": [
-        "! cat trainer/requirements.txt"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "e16cd8bb7483"
-      },
-      "outputs": [],
-      "source": [
-        "! pip install -r trainer/requirements.txt"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "0b8a210718c4"
-      },
-      "outputs": [],
-      "source": [
-        "! cat trainer/task.py"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "c0c6e7dfb3c6"
-      },
-      "outputs": [],
-      "source": [
-        "%run trainer/task.py --epochs 5 --no-cuda --local-mode"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "31dfdeede587"
-      },
-      "outputs": [],
-      "source": [
-        "! ls ./tmp"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "48d56ec621cc"
-      },
-      "outputs": [],
-      "source": [
-        "! rm -rf ./tmp"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "8f3ea1210749"
-      },
-      "source": [
-        "## Vertex Training using Vertex SDK and Custom Container"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "93002a20a2a6"
-      },
-      "source": [
-        "### Build Custom Container"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "4130ce43fd08"
-      },
-      "outputs": [],
-      "source": [
-        "hostname = \"gcr.io\"\n",
-        "image_name = content_name\n",
-        "tag = \"latest\"\n",
-        "\n",
-        "custom_container_image_uri = f\"{hostname}/{PROJECT_ID}/{image_name}:{tag}\""
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "2f1fc5b05240"
-      },
-      "outputs": [],
-      "source": [
-        "! cd trainer && docker build -t $custom_container_image_uri -f Dockerfile ."
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "b4f274f499ac"
-      },
-      "outputs": [],
-      "source": [
-        "! docker run --rm $custom_container_image_uri --epochs 5 --no-cuda --local-mode"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "ee1a0a06d0b4"
-      },
-      "outputs": [],
-      "source": [
-        "! docker push $custom_container_image_uri"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "cb763be12fc9"
-      },
-      "outputs": [],
-      "source": [
-        "! gcloud container images list --repository $hostname/$PROJECT_ID"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "10c8cc6b3334"
-      },
-      "source": [
-        "### Initialize Vertex SDK"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "1a12348169fa"
-      },
-      "outputs": [],
-      "source": [
-        "! pip install -r requirements.txt"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "42e981cefe41"
-      },
-      "outputs": [],
-      "source": [
-        "from google.cloud import aiplatform\n",
-        "\n",
-        "aiplatform.init(\n",
-        "    project=PROJECT_ID,\n",
-        "    staging_bucket=BUCKET_NAME,\n",
-        "    location=REGION,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "73c92c9298e9"
-      },
-      "source": [
-        "### Create a Vertex Tensorboard Instance"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "bde509558cd5"
-      },
-      "outputs": [],
-      "source": [
-        "content_name = content_name + \"-cpu\""
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "6d7908c0083c"
-      },
-      "outputs": [],
-      "source": [
-        "tensorboard = aiplatform.Tensorboard.create(\n",
-        "    display_name=content_name,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "a1f0a4f54037"
-      },
-      "source": [
-        "#### Option: Use a Previously Created Vertex Tensorboard Instance\n",
-        "\n",
-        "```\n",
-        "tensorboard_name = \"Your Tensorboard Resource Name or Tensorboard ID\"\n",
-        "tensorboard = aiplatform.Tensorboard(tensorboard_name=tensorboard_name)\n",
-        "```"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "a4cac84e04ac"
-      },
-      "source": [
-        "### Run a Vertex SDK CustomContainerTrainingJob"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "f92e8fdd44ee"
-      },
-      "outputs": [],
-      "source": [
-        "display_name = content_name\n",
-        "gcs_output_uri_prefix = f\"{BUCKET_NAME}/{display_name}\"\n",
-        "\n",
-        "replica_count = 4\n",
-        "machine_type = \"n1-standard-4\"\n",
-        "\n",
-        "args = [\n",
-        "    \"--backend\",\n",
-        "    \"gloo\",\n",
-        "    \"--no-cuda\",\n",
-        "    \"--batch-size\",\n",
-        "    \"128\",\n",
-        "    \"--epochs\",\n",
-        "    \"25\",\n",
-        "]"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "ae4c57df7e07"
-      },
-      "outputs": [],
-      "source": [
-        "custom_container_training_job = aiplatform.CustomContainerTrainingJob(\n",
-        "    display_name=display_name,\n",
-        "    container_uri=custom_container_image_uri,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "35cf3ecdf0df"
-      },
-      "outputs": [],
-      "source": [
-        "custom_container_training_job.run(\n",
-        "    args=args,\n",
-        "    base_output_dir=gcs_output_uri_prefix,\n",
-        "    replica_count=replica_count,\n",
-        "    machine_type=machine_type,\n",
-        "    tensorboard=tensorboard.resource_name,\n",
-        "    service_account=SERVICE_ACCOUNT,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "49d10dded73b"
-      },
-      "outputs": [],
-      "source": [
-        "print(f\"Custom Training Job Name: {custom_container_training_job.resource_name}\")\n",
-        "print(f\"GCS Output URI Prefix: {gcs_output_uri_prefix}\")"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "78398f52807b"
-      },
-      "source": [
-        "### Training Output Artifact"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "fc74422de1d1"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil ls $gcs_output_uri_prefix"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "5e99a6a05b10"
-      },
-      "source": [
-        "## Clean Up Artifact"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "b0c1b3f7466b"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil rm -rf $gcs_output_uri_prefix"
-      ]
-    }
-  ],
-  "metadata": {
-    "colab": {
-      "name": "multi_node_ddp_gloo_vertex_training_with_custom_container.ipynb",
-      "toc_visible": true
-    },
-    "kernelspec": {
-      "display_name": "Python 3",
-      "name": "python3"
-    }
-  },
-  "nbformat": 4,
-  "nbformat_minor": 0
-}

community-content/pytorch_image_classification_distributed_data_parallel_training_with_vertex_sdk/multi_node_ddp_nccl_vertex_training_with_custom_container.ipynb

@@ -1,347 +0,0 @@
-{
-  "cells": [
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "a6b56b1c7b76"
-      },
-      "outputs": [],
-      "source": [
-        "# Copyright 2021 Google LLC\n",
-        "#\n",
-        "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
-        "# you may not use this file except in compliance with the License.\n",
-        "# You may obtain a copy of the License at\n",
-        "#\n",
-        "#     https://www.apache.org/licenses/LICENSE-2.0\n",
-        "#\n",
-        "# Unless required by applicable law or agreed to in writing, software\n",
-        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
-        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
-        "# See the License for the specific language governing permissions and\n",
-        "# limitations under the License."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "20a5ea0081d0"
-      },
-      "source": [
-        "# PyTorch Image Classification Multi-Node Distributed Data Parallel Training on GPU using Vertex Training with Custom Container"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "8752d4a255fb"
-      },
-      "source": [
-        "<table align=\"left\">\n",
-        "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/master/community-content/pytorch_image_classification_distributed_data_parallel_training_with_vertex_sdk/multi_node_ddp_nccl_vertex_training_with_custom_container.ipynb\">\n",
-        "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
-        "      View on GitHub\n",
-        "    </a>\n",
-        "  </td>\n",
-        "</table>"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "03d216c7f7b1"
-      },
-      "source": [
-        "## Setup"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "c5ac73516218"
-      },
-      "outputs": [],
-      "source": [
-        "PROJECT_ID = \"YOUR PROJECT ID\"\n",
-        "BUCKET_NAME = \"gs://YOUR BUCKET NAME\"\n",
-        "REGION = \"YOUR REGION\"\n",
-        "SERVICE_ACCOUNT = \"YOUR SERVICE ACCOUNT\""
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "0b5ae674177e"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil ls -al $BUCKET_NAME"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "19a9b3bdd553"
-      },
-      "outputs": [],
-      "source": [
-        "content_name = \"pt-img-cls-multi-node-ddp-cust-cont\""
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "5307fe28b633"
-      },
-      "source": [
-        "## Vertex Training using Vertex SDK and Custom Container"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "46cb58c7fbf9"
-      },
-      "source": [
-        "### Built Custom Container"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "97e66e9f9bab"
-      },
-      "outputs": [],
-      "source": [
-        "hostname = \"gcr.io\"\n",
-        "image_name = content_name\n",
-        "tag = \"latest\"\n",
-        "\n",
-        "custom_container_image_uri = f\"{hostname}/{PROJECT_ID}/{image_name}:{tag}\""
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "ae9b29c4773f"
-      },
-      "source": [
-        "### Initialize Vertex SDK"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "dc1e84d5dec2"
-      },
-      "outputs": [],
-      "source": [
-        "! pip install -r requirements.txt"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "6964be27b98e"
-      },
-      "outputs": [],
-      "source": [
-        "from google.cloud import aiplatform\n",
-        "\n",
-        "aiplatform.init(\n",
-        "    project=PROJECT_ID,\n",
-        "    staging_bucket=BUCKET_NAME,\n",
-        "    location=REGION,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "594a91f438f2"
-      },
-      "source": [
-        "### Create a Vertex Tensorboard Instance"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "93134273261e"
-      },
-      "outputs": [],
-      "source": [
-        "content_name = content_name + \"-gpu\""
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "c2bd82dbcd9b"
-      },
-      "outputs": [],
-      "source": [
-        "tensorboard = aiplatform.Tensorboard.create(\n",
-        "    display_name=content_name,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "ebc593c6472e"
-      },
-      "source": [
-        "#### Option: Use a Previously Created Vertex Tensorboard Instance\n",
-        "\n",
-        "```\n",
-        "tensorboard_name = \"Your Tensorboard Resource Name or Tensorboard ID\"\n",
-        "tensorboard = aiplatform.Tensorboard(tensorboard_name=tensorboard_name)\n",
-        "```"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "0769e8e34c2f"
-      },
-      "source": [
-        "### Run a Vertex SDK CustomContainerTrainingJob"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "023f33ece826"
-      },
-      "outputs": [],
-      "source": [
-        "display_name = content_name\n",
-        "gcs_output_uri_prefix = f\"{BUCKET_NAME}/{display_name}\"\n",
-        "\n",
-        "replica_count = 1\n",
-        "machine_type = \"n1-standard-4\"\n",
-        "accelerator_count = 4\n",
-        "accelerator_type = \"NVIDIA_TESLA_K80\"\n",
-        "\n",
-        "args = [\n",
-        "    \"--backend\",\n",
-        "    \"nccl\",\n",
-        "    \"--batch-size\",\n",
-        "    \"128\",\n",
-        "    \"--epochs\",\n",
-        "    \"25\",\n",
-        "]"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "d4b599e726ef"
-      },
-      "outputs": [],
-      "source": [
-        "custom_container_training_job = aiplatform.CustomContainerTrainingJob(\n",
-        "    display_name=display_name,\n",
-        "    container_uri=custom_container_image_uri,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "81321e3bdf7f"
-      },
-      "outputs": [],
-      "source": [
-        "custom_container_training_job.run(\n",
-        "    args=args,\n",
-        "    base_output_dir=gcs_output_uri_prefix,\n",
-        "    replica_count=replica_count,\n",
-        "    machine_type=machine_type,\n",
-        "    accelerator_count=accelerator_count,\n",
-        "    accelerator_type=accelerator_type,\n",
-        "    tensorboard=tensorboard.resource_name,\n",
-        "    service_account=SERVICE_ACCOUNT,\n",
-        ")"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "5100712c2c4c"
-      },
-      "outputs": [],
-      "source": [
-        "print(f\"Custom Training Job Name: {custom_container_training_job.resource_name}\")\n",
-        "print(f\"GCS Output URI Prefix: {gcs_output_uri_prefix}\")"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "f9b77676e5a6"
-      },
-      "source": [
-        "### Training Output Artifact"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "0e171ce95ace"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil ls $gcs_output_uri_prefix"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "cf1b74a12b87"
-      },
-      "source": [
-        "## Clean Up Artifact"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "a0b15089c341"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil rm -rf $gcs_output_uri_prefix"
-      ]
-    }
-  ],
-  "metadata": {
-    "colab": {
-      "name": "multi_node_ddp_nccl_vertex_training_with_custom_container.ipynb",
-      "toc_visible": true
-    },
-    "kernelspec": {
-      "display_name": "Python 3",
-      "name": "python3"
-    }
-  },
-  "nbformat": 4,
-  "nbformat_minor": 0
-}

community-content/pytorch_pre_built_images_deployment/README.md

@@ -0,0 +1,30 @@
+# PyTorch Deployment on Google Cloud: Text Classification
+
+**This is an Experimental release**, covered by the Pre-GA Offerings Terms of your Google Cloud Platform [Terms of Service](https://cloud.google.com/terms).
+
+Experiments are focused on validating a prototype and are not guaranteed to be released. They are not intended for production use or covered by any SLA, support obligation, or deprecation policy and might be subject to backward-incompatible changes.
+
+**Kindly drop us a note before you run any scale tests.**
+
+**Do not hesitate to contact vertexai-prediction-preview-feedback@google.com if you have any questions or run into any issues.**
+
+The projects need to be added to the allowlist in order to deploy PyTorch models using Vertex AI Prediction pre-built PyTorch images. If you are interested in the feature, please send an email to vertexai-prediction-preview-feedback@google.com to provide your project numbers OR project ids.
+
+## Overview
+
+In the PyTorch on Google Cloud series of blog posts, we aim to share how to deploy PyTorch models at scale on [Vertex AI](https://cloud.google.com/vertex-ai).
+
+This tutorial on text classification shows how to deploy a PyTorch based text classification model on [Vertex AI](https://cloud.google.com/vertex-ai/docs/start/client-libraries#python) using Vertex SDK and [`gcloud ai`](https://cloud.google.com/sdk/gcloud/reference/beta/ai).
+
+## Notebooks
+
+| <h4>Notebook</h4> | <h4>Description</h4>                       |
+| :-------- | :------- |
+| [pytorch-text-classification-vertex-ai-deploy.ipynb](./pytorch-text-classification-vertex-ai-deploy.ipynb) | Notebook to show deploying a PyTorch model on Vertex AI |
+
+## Folders
+
+
+| <h4>Folder Name</h4> | <h4>Description</h4>                       |
+| :-------- | :------- |
+| [`predictor`](./predictor) | Folder with custom prediction handler to deploy a PyTorch model to Vertex Prediction. In the [notebook](./pytorch-text-classification-vertex-ai-deploy.ipynb), this folder is used for deploying a PyTorch model on Vertex AI using Vertex Prediction pre-built PyTorch images |

community-content/pytorch_pre_built_images_deployment/predictor/custom_handler.py

@@ -0,0 +1,91 @@
+
+import os
+import json
+import logging
+
+import torch
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+from ts.torch_handler.base_handler import BaseHandler
+
+logger = logging.getLogger(__name__)
+
+
+class TransformersClassifierHandler(BaseHandler):
+    """
+    The handler takes an input string and returns the classification text 
+    based on the serialized transformers checkpoint.
+    """
+    def __init__(self):
+        super(TransformersClassifierHandler, self).__init__()
+        self.initialized = False
+
+    def initialize(self, ctx):
+        """ Loads the model.pt file and initialized the model object.
+        Instantiates Tokenizer for preprocessor to use
+        Loads labels to name mapping file for post-processing inference response
+        """
+        self.manifest = ctx.manifest
+
+        properties = ctx.system_properties
+        model_dir = properties.get("model_dir")
+        self.device = torch.device("cuda:" + str(properties.get("gpu_id")) if torch.cuda.is_available() else "cpu")
+
+        # Read model serialize/pt file
+        serialized_file = self.manifest["model"]["serializedFile"]
+        model_pt_path = os.path.join(model_dir, serialized_file)
+        if not os.path.isfile(model_pt_path):
+            raise RuntimeError("Missing the model.pt or pytorch_model.bin file")
+        
+        # Load model
+        self.model = AutoModelForSequenceClassification.from_pretrained(model_dir)
+        self.model.to(self.device)
+        self.model.eval()
+        logger.debug('Transformer model from path {0} loaded successfully'.format(model_dir))
+        
+        # Ensure to use the same tokenizer used during training
+        self.tokenizer = AutoTokenizer.from_pretrained('bert-base-cased')
+
+        # Read the mapping file, index to object name
+        mapping_file_path = os.path.join(model_dir, "index_to_name.json")
+
+        if os.path.isfile(mapping_file_path):
+            with open(mapping_file_path) as f:
+                self.mapping = json.load(f)
+        else:
+            logger.warning('Missing the index_to_name.json file. Inference output will default.')
+            self.mapping = {"0": "Negative",  "1": "Positive"}
+
+        self.initialized = True
+
+    def preprocess(self, data):
+        """ Preprocessing input request by tokenizing
+            Extend with your own preprocessing steps as needed
+        """
+        text = data[0].get("data")
+        if text is None:
+            text = data[0].get("body")
+        sentences = text.decode('utf-8')
+        logger.info("Received text: '%s'", sentences)
+
+        # Tokenize the texts
+        tokenizer_args = ((sentences,))
+        inputs = self.tokenizer(*tokenizer_args,
+                                padding='max_length',
+                                max_length=128,
+                                truncation=True,
+                                return_tensors = "pt")
+        return inputs
+
+    def inference(self, inputs):
+        """ Predict the class of a text using a trained transformer model.
+        """
+        prediction = self.model(inputs['input_ids'].to(self.device))[0].argmax().item()
+
+        if self.mapping:
+            prediction = self.mapping[str(prediction)]
+
+        logger.info("Model predicted: '%s'", prediction)
+        return [prediction]
+
+    def postprocess(self, inference_output):
+        return inference_output

community-content/pytorch_pre_built_images_deployment/predictor/index_to_name.json

@@ -0,0 +1,5 @@
+
+{
+    "0": "Negative", 
+    "1": "Positive"
+}

community-content/pytorch_text_classification_using_vertex_sdk_and_gcloud/pytorch-text-classification-vertex-ai-train-tune-deploy.ipynb

@@ -658,8 +658,8 @@
       },
       "outputs": [],
       "source": [
-        "datasets = load_dataset(\"imdb\")\n",
-        "datasets"
+        "dataset = load_dataset(\"imdb\")\n",
+        "dataset"
       ]
     },
     {
@@ -668,7 +668,7 @@
         "id": "RzfPtOMoIrIu"
       },
       "source": [
-        "The `datasets` object itself is [`DatasetDict`](https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasetdict), which contains one key for the training, validation and test set."
+        "The `dataset` object itself is [`DatasetDict`](https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasetdict), which contains one key for the training, validation and test set."
       ]
     },
     {
@@ -681,12 +681,12 @@
       "source": [
         "print(\n",
         "    \"Total # of rows in training dataset {} and size {:5.2f} MB\".format(\n",
-        "        datasets[\"train\"].shape[0], datasets[\"train\"].size_in_bytes / (1024 * 1024)\n",
+        "        dataset[\"train\"].shape[0], dataset[\"train\"].size_in_bytes / (1024 * 1024)\n",
         "    )\n",
         ")\n",
         "print(\n",
         "    \"Total # of rows in test dataset {} and size {:5.2f} MB\".format(\n",
-        "        datasets[\"test\"].shape[0], datasets[\"test\"].size_in_bytes / (1024 * 1024)\n",
+        "        dataset[\"test\"].shape[0], dataset[\"test\"].size_in_bytes / (1024 * 1024)\n",
         "    )\n",
         ")"
       ]
@@ -708,7 +708,7 @@
       },
       "outputs": [],
       "source": [
-        "datasets[\"train\"][0]"
+        "dataset[\"train\"][0]"
       ]
     },
     {
@@ -728,7 +728,7 @@
       },
       "outputs": [],
       "source": [
-        "label_list = datasets[\"train\"].unique(\"label\")\n",
+        "label_list = dataset[\"train\"].unique(\"label\")\n",
         "label_list"
       ]
     },
@@ -779,7 +779,7 @@
       },
       "outputs": [],
       "source": [
-        "show_random_elements(datasets[\"train\"])"
+        "show_random_elements(dataset[\"train\"])"
       ]
     },
     {
@@ -883,7 +883,7 @@
       },
       "outputs": [],
       "source": [
-        "example = datasets[\"train\"][4]\n",
+        "example = dataset[\"train\"][4]\n",
         "print(example)"
       ]
     },
@@ -920,7 +920,7 @@
       "source": [
         "# Dataset loading repeated here to make this cell idempotent\n",
         "# Since we are over-writing datasets variable\n",
-        "datasets = load_dataset(\"imdb\")\n",
+        "dataset = load_dataset(\"imdb\")\n",
         "\n",
         "# Mapping labels to ids\n",
         "# NOTE: We can extract this automatically but the `Unique` method of the datasets\n",
@@ -948,7 +948,7 @@
         "\n",
         "\n",
         "# apply preprocessing function to input examples\n",
-        "datasets = datasets.map(preprocess_function, batched=True, load_from_cache_file=True)"
+        "dataset = dataset.map(preprocess_function, batched=True, load_from_cache_file=True)"
       ]
     },
     {
@@ -1091,8 +1091,8 @@
         "trainer = Trainer(\n",
         "    model,\n",
         "    args,\n",
-        "    train_dataset=datasets[\"train\"],\n",
-        "    eval_dataset=datasets[\"test\"],\n",
+        "    train_dataset=dataset[\"train\"],\n",
+        "    eval_dataset=dataset[\"test\"],\n",
         "    data_collator=default_data_collator,\n",
         "    tokenizer=tokenizer,\n",
         "    compute_metrics=compute_metrics,\n",

community-content/tf_agents_bandits_movie_recommendation_with_kfp_and_vertex_sdk/mlops_pipeline_tf_agents_bandits_movie_recommendation/src/logger/requirements.txt

@@ -1,4 +1,4 @@
 google-cloud-bigquery==2.20.0
-tensorflow==2.5.3
+tensorflow==2.7.2
 pillow==9.0.1
 tf-agents==0.8.0

community-content/tf_keras_image_classification_distributed_multi_worker_with_vertex_sdk/trainer/requirements.txt

@@ -1 +1 @@
-tensorflow==2.5.3
+tensorflow==2.7.2

notebooks/README.md

@@ -1,5 +1,5 @@
-The [official](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/official) folder contains notebooks organized by Google Cloud product.
+The [official](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/official) folder contains notebooks organized by Google Cloud product. These are tested weekly and maintained by Google.
 
-The [community](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community) folder contains notebooks that aren't officially supported by Google.
+The [community](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community) folder contains notebooks that may be created by Google or external contributors. They are not necessary maintained.
 
 Contributions to the repo should use the [notebook template](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/notebook_template.ipynb) as a starting point.

notebooks/community/CODEOWNERS

@@ -5,11 +5,13 @@
 
 /sdk/sdk_* @andrewferlitsch
 /gapic @andrewferlitsch
+/gapic/custom/showcase_custom_image_classification_online_explain_example_based_api.ipynb @inardini
 /ml_ops @andrewferlitsch
 /model_monitoring/* @mco-gh
 /structured_data/rapid_prototyping_* @rafael-carvalho
 
 /managed_notebooks/
+/bigquery_ml/ @polong
 /sdk/SDK_FBProphet_Forecasting_Online.ipynb @brianchunkang
 /pipelines/google_cloud_pipeline_components_TPU_model_train_upload_deploy.ipynb @brianchunkang
 /explainable_ai/SDK_Custom_Container_XAI.ipynb @brianchunkang
@@ -18,7 +20,11 @@
 /sdk/pytorch_lightning_custom_container_training.ipynb @brianchunkang
 /tensorboard @yfang1
 /feature_store @nayaknishant @morgandu
+/prediction @googleapis/vertex-prediction-team
 /vertex_endpoints/tf_hub_obj_detection/deploy_tfhub_object_detection_on_vertex_endpoints.ipynb @entrpn
 /vertex_endpoints/nvidia-triton/nvidia-triton-custom-container-prediction.ipynb @RajeshThallam
 /vertex_endpoints/optimized_tensorflow_runtime @vlasenkoalexey
-/notebooks/community/ml_ops/stage2/get_started_with_visionapi_and_automl.ipynb @mansari
+/notebooks/community/ml_ops/stage2/get_started_with_visionapi_and_automl.ipynb @mansari
+/notebooks/community/neo4j/graph_paysim.ipynb @benofben @laeg
+/notebooks/community/ml_ops/stage1/get_started_with_visionapi_and_vertex_datasets.ipynb @mansari
+/notebooks/community/pipelines/google_cloud_pipeline_components_bqml_pipeline_demand_forecasting.ipynb @inardini

notebooks/community/feature_store/gapic-feature-store.ipynb

@@ -292,6 +292,37 @@
         "    PROJECT_ID = \"python-docs-samples-tests\"  # @param {type:\"string\"}"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "d9f118b92c74"
+      },
+      "source": [
+        "#### UUID\n",
+        "\n",
+        "If you are in a live tutorial session, you might be using a shared test account or project. To avoid name collisions between users on resources created, you create a uuid for each instance session, and append it onto the name of resources you create in this tutorial."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "3ee72715c0fd"
+      },
+      "outputs": [],
+      "source": [
+        "import random\n",
+        "import string\n",
+        "\n",
+        "\n",
+        "# Generate a uuid of a specifed length(default=8)\n",
+        "def generate_uuid(length: int = 8) -> str:\n",
+        "    return \"\".join(random.choices(string.ascii_lowercase + string.digits, k=length))\n",
+        "\n",
+        "\n",
+        "UUID = generate_uuid()"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -478,7 +509,6 @@
         "from google.cloud.aiplatform_v1.types import \\\n",
         "    featurestore_service as featurestore_service_pb2\n",
         "from google.cloud.aiplatform_v1.types import io as io_pb2\n",
-        "from google.protobuf.duration_pb2 import Duration\n",
         "\n",
         "# Create admin_client for CRUD and data_client for reading feature values.\n",
         "admin_client = FeaturestoreServiceClient(client_options={\"api_endpoint\": API_ENDPOINT})\n",
@@ -542,7 +572,7 @@
       },
       "outputs": [],
       "source": [
-        "FEATURESTORE_ID = \"movie_prediction\"\n",
+        "FEATURESTORE_ID = f\"movie_prediction_{UUID}\"\n",
         "try:\n",
         "    create_lro = admin_client.create_featurestore(\n",
         "        featurestore_service_pb2.CreateFeaturestoreRequest(\n",
@@ -567,7 +597,7 @@
         "id": "ag8pCQ7rNjVf"
       },
       "source": [
-        "You can use [GetFeaturestore](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1beta1#google.cloud.aiplatform.v1beta1.FeaturestoreService.GetFeaturestore) or [ListFeaturestores](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1beta1#google.cloud.aiplatform.v1beta1.FeaturestoreService.ListFeaturestores) to check if the Featurestore was successfully created. The following example gets the details of the Featurestore.\n"
+        "You can use [GetFeaturestore](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1#google.cloud.aiplatform.v1.FeaturestoreService.GetFeaturestore) or [ListFeaturestores](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1#google.cloud.aiplatform.v1.FeaturestoreService.ListFeaturestores) to check if the Featurestore was successfully created. The following example gets the details of the Featurestore.\n"
       ]
     },
     {
@@ -589,7 +619,7 @@
         "id": "018ab19d934f"
       },
       "source": [
-        "Auto scaling is available in v1beta1 since v1.11. Below is the example for the `CreateFeaturestoreRequest` with auto-scaling, use it with `aiplatform_v1beta1.FeaturestoreServiceClient` to create Featurestore:"
+        "Auto scaling is available in v1 since v1.11. Below is the example for the `CreateFeaturestoreRequest` with auto-scaling, use it with `aiplatform_v1.FeaturestoreServiceClient` to create Featurestore:"
       ]
     },
     {
@@ -600,17 +630,17 @@
       },
       "outputs": [],
       "source": [
-        "from google.cloud.aiplatform_v1beta1.types import \\\n",
-        "    featurestore as v1beta1_featurestore_pb2\n",
-        "from google.cloud.aiplatform_v1beta1.types import \\\n",
-        "    featurestore_service as v1beta1_featurestore_service_pb2\n",
+        "from google.cloud.aiplatform_v1.types import \\\n",
+        "    featurestore as v1_featurestore_pb2\n",
+        "from google.cloud.aiplatform_v1.types import \\\n",
+        "    featurestore_service as v1_featurestore_service_pb2\n",
         "\n",
-        "create_featurestore_request = v1beta1_featurestore_service_pb2.CreateFeaturestoreRequest(\n",
+        "create_featurestore_request = v1_featurestore_service_pb2.CreateFeaturestoreRequest(\n",
         "    parent=BASE_RESOURCE_PATH,\n",
         "    featurestore_id=FEATURESTORE_ID,\n",
-        "    featurestore=v1beta1_featurestore_pb2.Featurestore(\n",
-        "        online_serving_config=v1beta1_featurestore_pb2.Featurestore.OnlineServingConfig(\n",
-        "            scaling=v1beta1_featurestore_pb2.Featurestore.OnlineServingConfig.Scaling(\n",
+        "    featurestore=v1_featurestore_pb2.Featurestore(\n",
+        "        online_serving_config=v1_featurestore_pb2.Featurestore.OnlineServingConfig(\n",
+        "            scaling=v1_featurestore_pb2.Featurestore.OnlineServingConfig.Scaling(\n",
         "                min_node_count=1, max_node_count=5\n",
         "            )\n",
         "        ),\n",
@@ -675,14 +705,67 @@
         "    print(e)"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dPkT7KDuEvWv"
+      },
+      "source": [
+        "Feature [monitoring](https://cloud.google.com/vertex-ai/docs/featurestore/monitoring) is in preview, so you need to use v1 Python. Import feature analysis is only available through SDK for now."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "9kiqrBN6E28r"
+      },
+      "outputs": [],
+      "source": [
+        "from google.cloud.aiplatform_v1 import \\\n",
+        "    FeaturestoreServiceClient as v1_FeaturestoreServiceClient\n",
+        "from google.cloud.aiplatform_v1.types import entity_type as v1_entity_type_pb2\n",
+        "from google.cloud.aiplatform_v1.types import \\\n",
+        "    featurestore_monitoring as v1_featurestore_monitoring_pb2\n",
+        "from google.cloud.aiplatform_v1.types import \\\n",
+        "    featurestore_service as v1_featurestore_service_pb2\n",
+        "\n",
+        "v1_admin_client = v1_FeaturestoreServiceClient(\n",
+        "    client_options={\"api_endpoint\": API_ENDPOINT}\n",
+        ")\n",
+        "\n",
+        "# Enable import feature analysis for users entity type.\n",
+        "# All Features belonging to this EntityType will by default inherit the monitoring config.\n",
+        "v1_admin_client.update_entity_type(\n",
+        "    v1_featurestore_service_pb2.UpdateEntityTypeRequest(\n",
+        "        entity_type=v1_entity_type_pb2.EntityType(\n",
+        "            name=admin_client.entity_type_path(\n",
+        "                PROJECT_ID, REGION, FEATURESTORE_ID, \"users\"\n",
+        "            ),\n",
+        "            monitoring_config=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig(\n",
+        "                import_features_analysis=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ImportFeaturesAnalysis(\n",
+        "                    anomaly_detection_baseline=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ImportFeaturesAnalysis.Baseline.LATEST_STATS,\n",
+        "                    state=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ImportFeaturesAnalysis.State.ENABLED,\n",
+        "                ),\n",
+        "                numerical_threshold_config=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ThresholdConfig(\n",
+        "                    value=0.001,\n",
+        "                ),\n",
+        "                categorical_threshold_config=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ThresholdConfig(\n",
+        "                    value=0.001,\n",
+        "                ),\n",
+        "            ),\n",
+        "        ),\n",
+        "    )\n",
+        ")"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "85b1f59fbf6d"
       },
       "source": [
-        "Feature [monitoring](https://cloud.google.com/vertex-ai/docs/featurestore/monitoring) is in preview, so you need to use v1beta1 Python. The easiest way to set this for now is using [console UI](https://console.cloud.google.com/vertex-ai/features). For completeness, below is example to do this using v1beta1 SDK.\n",
-        "\n",
+        "The easiest way to set up snapshot analysis for now is using [console UI](https://console.cloud.google.com/vertex-ai/features). For completeness, below is example to do this using v1 SDK.\n",
         "\n",
         "You can view monitoring statistics on [console UI](https://console.cloud.google.com/vertex-ai/features)."
       ]
@@ -695,30 +778,36 @@
       },
       "outputs": [],
       "source": [
-        "from google.cloud.aiplatform_v1beta1 import \\\n",
-        "    FeaturestoreServiceClient as v1beta1_FeaturestoreServiceClient\n",
-        "from google.cloud.aiplatform_v1beta1.types import \\\n",
-        "    entity_type as v1beta1_entity_type_pb2\n",
-        "from google.cloud.aiplatform_v1beta1.types import \\\n",
-        "    featurestore_monitoring as v1beta1_featurestore_monitoring_pb2\n",
-        "from google.cloud.aiplatform_v1beta1.types import \\\n",
-        "    featurestore_service as v1beta1_featurestore_service_pb2\n",
+        "from google.cloud.aiplatform_v1 import \\\n",
+        "    FeaturestoreServiceClient as v1_FeaturestoreServiceClient\n",
+        "from google.cloud.aiplatform_v1.types import entity_type as v1_entity_type_pb2\n",
+        "from google.cloud.aiplatform_v1.types import \\\n",
+        "    featurestore_monitoring as v1_featurestore_monitoring_pb2\n",
+        "from google.cloud.aiplatform_v1.types import \\\n",
+        "    featurestore_service as v1_featurestore_service_pb2\n",
         "\n",
-        "v1beta1_admin_client = v1beta1_FeaturestoreServiceClient(\n",
+        "v1_admin_client = v1_FeaturestoreServiceClient(\n",
         "    client_options={\"api_endpoint\": API_ENDPOINT}\n",
         ")\n",
         "\n",
-        "# Enable monitoring for users entity type.\n",
+        "# Enable snapshot analysis for users entity type.\n",
         "# All Features belonging to this EntityType will by default inherit the monitoring config.\n",
-        "v1beta1_admin_client.update_entity_type(\n",
-        "    v1beta1_featurestore_service_pb2.UpdateEntityTypeRequest(\n",
-        "        entity_type=v1beta1_entity_type_pb2.EntityType(\n",
+        "v1_admin_client.update_entity_type(\n",
+        "    v1_featurestore_service_pb2.UpdateEntityTypeRequest(\n",
+        "        entity_type=v1_entity_type_pb2.EntityType(\n",
         "            name=admin_client.entity_type_path(\n",
         "                PROJECT_ID, REGION, FEATURESTORE_ID, \"users\"\n",
         "            ),\n",
-        "            monitoring_config=v1beta1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig(\n",
-        "                snapshot_analysis=v1beta1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.SnapshotAnalysis(\n",
-        "                    monitoring_interval=Duration(seconds=86400),  # 1 day\n",
+        "            monitoring_config=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig(\n",
+        "                snapshot_analysis=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.SnapshotAnalysis(\n",
+        "                    monitoring_interval_days=1,  # 1 day\n",
+        "                    staleness_days=30,\n",
+        "                ),\n",
+        "                numerical_threshold_config=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ThresholdConfig(\n",
+        "                    value=0.001,\n",
+        "                ),\n",
+        "                categorical_threshold_config=v1_featurestore_monitoring_pb2.FeaturestoreMonitoringConfig.ThresholdConfig(\n",
+        "                    value=0.001,\n",
         "                ),\n",
         "            ),\n",
         "        ),\n",
@@ -754,6 +843,7 @@
         "                feature=feature_pb2.Feature(\n",
         "                    value_type=feature_pb2.Feature.ValueType.INT64,\n",
         "                    description=\"User age\",\n",
+        "                    disable_monitoring=False,\n",
         "                ),\n",
         "                feature_id=\"age\",\n",
         "            ),\n",
@@ -761,6 +851,8 @@
         "                feature=feature_pb2.Feature(\n",
         "                    value_type=feature_pb2.Feature.ValueType.STRING,\n",
         "                    description=\"User gender\",\n",
+        "                    # Default is False. If True, Feature 'gender' monitoring analysis is disabled.\n",
+        "                    disable_monitoring=True,\n",
         "                ),\n",
         "                feature_id=\"gender\",\n",
         "            ),\n",
@@ -827,8 +919,8 @@
       "source": [
         "## Search created features\n",
         "\n",
-        "While the [ListFeatures](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1beta1#google.cloud.aiplatform.v1beta1.FeaturestoreService.ListFeatures) method allows you to easily view all features of a single\n",
-        "entity type, the [SearchFeatures](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1beta1#google.cloud.aiplatform.v1beta1.FeaturestoreService.SearchFeatures) method searches across all featurestores\n",
+        "While the [ListFeatures](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1#google.cloud.aiplatform.v1.FeaturestoreService.ListFeatures) method allows you to easily view all features of a single\n",
+        "entity type, the [SearchFeatures](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1#google.cloud.aiplatform.v1.FeaturestoreService.SearchFeatures) method searches across all featurestores\n",
         "and entity types in a given location (such as `us-central1`). This can help you discover features that were created by someone else.\n",
         "\n",
         "You can query based on feature properties including feature ID, entity type ID,\n",
@@ -1030,6 +1122,8 @@
         "    ],\n",
         "    feature_time_field=\"update_time\",\n",
         "    worker_count=1,\n",
+        "    # Default is False. If True, the import feature analysis won't happen for this specific operation.\n",
+        "    disable_ingestion_analysis=False,\n",
         ")"
       ]
     },
@@ -1140,7 +1234,7 @@
       },
       "source": [
         "The\n",
-        "[Online Serving APIs](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1beta1#featurestoreonlineservingservice)\n",
+        "[Online Serving APIs](https://cloud.google.com/vertex-ai/docs/reference/rpc/google.cloud.aiplatform.v1#featurestoreonlineservingservice)\n",
         "lets you serve feature values for small batches of entities. It's designed for latency-sensitive service, such as online model prediction. For example, for a movie service, you might want to quickly shows movies that the current user would most likely watch by using online predictions."
       ]
     },
@@ -1438,9 +1532,7 @@
   ],
   "metadata": {
     "colab": {
-      "collapsed_sections": [
-        "ze4-nDLfK4pw"
-      ],
+      "collapsed_sections": [],
       "name": "gapic-feature-store.ipynb",
       "toc_visible": true
     },

notebooks/community/feature_store/mobile_gaming/mobile_gaming_feature_store.ipynb

@@ -32,18 +32,18 @@
         "<table align=\"left\">\n",
         "\n",
         "  <td>\n",
-        "    <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/notebook_template.ipynb\">\n",
+        "    <a href=\"https://colab.research.google.com/github/vertex-ai-samples/blob/main/notebooks/community/feature_store/mobile_gaming/mobile_gaming_feature_store.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/notebook_template.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/feature_store/mobile_gaming/mobile_gaming_feature_store.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/notebook_template.ipynb\">\n",
+        "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/feature_store/mobile_gaming/mobile_gaming_feature_store.ipynb\">\n",
         "      <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
         "      Open in Vertex AI Workbench\n",
         "    </a>\n",
@@ -54,52 +54,52 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "7FZeBEwdXS4d"
+        "id": "tvgnzT1CKxrO"
       },
       "source": [
         "## Overview\n",
-        "\n",
-        "Imagine you are a member of the Data Science team working on the same Mobile Gaming application reported in the [Churn prediction for game developers using Google Analytics 4 (GA4) and BigQuery ML](https://cloud.google.com/blog/topics/developers-practitioners/churn-prediction-game-developers-using-google-analytics-4-ga4-and-bigquery-ml) blog post. \n",
-        "\n",
-        "Business wants to use that information in real-time to take immediate intervention actions in-game to prevent churn. In particular, for each player, they want to provide gaming incentives like new items or bonus packs depending on the customer demographic,  behavioral information and the resulting propensity of return. \n",
-        "\n",
-        "Last year, Google Cloud announced Vertex AI, a managed machine learning (ML) platform that allows data science teams to accelerate the deployment and maintenance of ML models. One of the platform building blocks is Vertex AI Feature store which provides a managed service for low latency scalable feature serving. Also it is a centralized feature repository with easy APIs to search & discover features and feature monitoring capabilities to track drift and other quality issues. \n",
-        "\n",
-        "In this notebook, we will show how the role of Vertex AI Feature Store in a ready to production scenario when the user's activities within the first 24 hours of last engagment and the gaming platform would consume in order to improver UX. Below you can find the high level picture of the system\n",
-        "\n",
+        " \n",
+        "Imagine you are a member of the Data Science team working on the same Mobile Gaming application reported in the [Churn prediction for game developers using Google Analytics 4 (GA4) and BigQuery ML](https://cloud.google.com/blog/topics/developers-practitioners/churn-prediction-game-developers-using-google-analytics-4-ga4-and-bigquery-ml) blog post.\n",
+        " \n",
+        "Business wants to use that information in real-time to take immediate intervention actions in-game to prevent churn. In particular, for each player, they want to provide gaming incentives like new items or bonus packs depending on the customer demographic,  behavioral information and the resulting propensity of return.\n",
+        " \n",
+        "Last year, Google Cloud announced Vertex AI, a managed machine learning (ML) platform that allows data science teams to accelerate the deployment and maintenance of ML models. One of the platform building blocks is Vertex AI Feature store which provides a managed service for low latency scalable feature serving. Also it is a centralized feature repository with easy APIs to search & discover features and feature monitoring capabilities to track drift and other quality issues.\n",
+        " \n",
+        "In this notebook, we will show how the role of Vertex AI Feature Store in a ready to production scenario when the user's activities within the first 24 hours of last engagement and the gaming platform would consume in order to improve UX. Below you can find the high level picture of the system\n",
+        " \n",
         "<img src=\"./assets/mobile_gaming_architecture_1.png\">\n",
-        "\n",
-        "\n",
+        " \n",
+        " \n",
         "### Dataset\n",
-        "\n",
+        " \n",
         "The dataset is the public sample export data from an actual mobile game app called \"Flood It!\" (Android, iOS)\n",
-        "\n",
+        " \n",
         "### Objective\n",
-        "\n",
+        " \n",
         "In the following notebook, you will learn how Vertex AI Feature store\n",
-        "\n",
-        "1.   Provide a centralized feature repository with easy APIs to search & discover features and fetch them for training/serving. \n",
-        "\n",
-        "2.   Simplify deployments of models for Online Prediction, via low latency scalable feature serving.\n",
-        "\n",
-        "3.   Mitigate training serving skew and data leakage by performing point in time lookups to fetch historical data for training.\n",
-        "\n",
+        " \n",
+        "1.  Provide a centralized feature repository with easy APIs to search & discover features and fetch them for training/serving.\n",
+        " \n",
+        "2.  Simplify deployments of models for Online Prediction, via low latency scalable feature serving.\n",
+        " \n",
+        "3.  Mitigate training serving skew and data leakage by performing point in time lookups to fetch historical data for training.\n",
+        " \n",
         "**Notice that we assume that already know how to set up a Vertex AI Feature store. In case you are not, please check out [this detailed notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/feature_store/gapic-feature-store.ipynb).**\n",
-        "\n",
-        "\n",
-        "### Costs \n",
-        "\n",
+        " \n",
+        " \n",
+        "### Costs\n",
+        " \n",
         "This tutorial uses billable components of Google Cloud:\n",
-        "\n",
+        " \n",
         "* Vertex AI\n",
         "* BigQuery\n",
         "* Cloud Storage\n",
-        "\n",
+        " \n",
         "Learn about [Vertex AI\n",
         "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
         "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
         "Calculator](https://cloud.google.com/products/calculator/)\n",
-        "to generate a cost estimate based on your projected usage."
+        "to generate a cost estimate based on your projected usage.\n"
       ]
     },
     {
@@ -110,7 +110,7 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "**If you are using Colab or Google Cloud Notebooks**, your environment already meets\n",
+        "**If you are using Colab or Vertex AI Workbench Notebooks**, your environment already meets\n",
         "all the requirements to run this notebook. You can skip this step."
       ]
     },
@@ -159,7 +159,7 @@
       "source": [
         "### Install additional packages\n",
         "\n",
-        "Install additional package dependencies not installed in your notebook environment, such as {XGBoost, AdaNet, or TensorFlow Hub TODO: Replace with relevant packages for the tutorial}. Use the latest major GA version of each package."
+        "Install additional package dependencies not installed in your notebook environment, such as XGBoost. Use the latest major GA version of each package."
       ]
     },
     {
@@ -172,12 +172,15 @@
       "source": [
         "import os\n",
         "\n",
-        "# The Google Cloud Notebook product has specific requirements\n",
-        "IS_GOOGLE_CLOUD_NOTEBOOK = os.path.exists(\"/opt/deeplearning/metadata/env_version\")\n",
+        "# The Vertex AI Workbench Notebook product has specific requirements\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
+        "    \"/opt/deeplearning/metadata/env_version\"\n",
+        ")\n",
         "\n",
-        "# Google Cloud Notebook requires dependencies to be installed with '--user'\n",
+        "# Vertex AI Notebook requires dependencies to be installed with '--user'\n",
         "USER_FLAG = \"\"\n",
-        "if IS_GOOGLE_CLOUD_NOTEBOOK:\n",
+        "if IS_WORKBENCH_NOTEBOOK:\n",
         "    USER_FLAG = \"--user\""
       ]
     },
@@ -185,11 +188,11 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "SzEo6DeE2GOP"
+        "id": "_vr6BYED_5my"
       },
       "outputs": [],
       "source": [
-        "! pip3 install {USER_FLAG} --upgrade pip\n",
+        "! pip3 install {USER_FLAG} --upgrade pip -q\n",
         "! pip3 install {USER_FLAG} --upgrade google-cloud-aiplatform==1.11.0 -q --no-warn-conflicts\n",
         "! pip3 install {USER_FLAG} git+https://github.com/googleapis/python-aiplatform.git@main # For features monitoring\n",
         "! pip3 install {USER_FLAG} --upgrade google-cloud-bigquery==2.24.0 -q --no-warn-conflicts\n",
@@ -249,7 +252,7 @@
         "\n",
         "1. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
         "\n",
-        "1. [Enable the Vertex AI API and Compute Engine API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component). \n",
+        "1. [Enable the APIs](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,notebooks.googleapis.com, ). \n",
         "\n",
         "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
@@ -307,18 +310,76 @@
       "outputs": [],
       "source": [
         "if PROJECT_ID == \"\" or PROJECT_ID is None:\n",
-        "    PROJECT_ID = \"\"  # @param {type:\"string\"}"
+        "    PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "dEjRdjxBuDsi"
+        "id": "set_gcloud_project_id"
       },
       "outputs": [],
       "source": [
-        "!gcloud config set project $PROJECT_ID #change it"
+        "! gcloud config set project $PROJECT_ID"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "23988890fef6"
+      },
+      "source": [
+        "#### Get your project number (Optional)\n",
+        "\n",
+        "Now that the project ID is set, you get your corresponding project number."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "2d6950574e1d"
+      },
+      "outputs": [],
+      "source": [
+        "shell_output = ! gcloud projects list --filter=\"PROJECT_ID:'{PROJECT_ID}'\" --format='value(PROJECT_NUMBER)'\n",
+        "PROJECT_NUMBER = shell_output[0]\n",
+        "print(\"Project Number:\", PROJECT_NUMBER)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "region"
+      },
+      "source": [
+        "#### Region\n",
+        "\n",
+        "You can also change the `REGION` variable, which is used for operations\n",
+        "throughout the rest of this notebook.  Below are regions supported for Vertex AI. We recommend that you choose the region closest to you.\n",
+        "\n",
+        "- Americas: `us-central1`\n",
+        "- Europe: `europe-west4`\n",
+        "- Asia Pacific: `asia-east1`\n",
+        "\n",
+        "You may not use a multi-regional bucket for training with Vertex AI. Not all regions provide support for all Vertex AI services.\n",
+        "\n",
+        "Learn more about [Vertex AI regions](https://cloud.google.com/vertex-ai/docs/general/locations)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "jIcZV7-C2RrX"
+      },
+      "outputs": [],
+      "source": [
+        "REGION = \"[your-region]\"  # @param {type: \"string\"}\n",
+        "\n",
+        "if REGION == \"[your-region]\":\n",
+        "    REGION = \"us-central1\""
       ]
     },
     {
@@ -353,7 +414,7 @@
       "source": [
         "### Authenticate your Google Cloud account\n",
         "\n",
-        "**If you are using Google Cloud Notebooks**, your environment is already\n",
+        "**If you are using Vertex AI Workbench Notebooks**, your environment is already\n",
         "authenticated. Skip this step."
       ]
     },
@@ -376,9 +437,13 @@
         "3. In the **Service account name** field, enter a name, and\n",
         "   click **Create**.\n",
         "\n",
-        "4. In the **Grant this service account access to project** section, click the **Role** drop-down list. Type \"Vertex AI\"\n",
-        "into the filter box, and select\n",
-        "   **Vertex AI Administrator**. Type \"Storage Object Admin\" into the filter box, and select **Storage Object Admin**.\n",
+        "4. In the **Grant this service account access to project** section, click the **Role** drop-down list and add the following roles:\n",
+        "   - BigQuery Admin\n",
+        "   - Storage Admin\n",
+        "   - Storage Object Admin\n",
+        "   - Vertex AI Administrator\n",
+        "   - Vertex AI Feature Store Admin\n",
+        "\n",
         "\n",
         "5. Click *Create*. A JSON file that contains your key downloads to your\n",
         "local environment.\n",
@@ -395,19 +460,19 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "import sys\n",
-        "\n",
         "# If you are running this notebook in Colab, run this cell and follow the\n",
         "# instructions to authenticate your GCP account. This provides access to your\n",
         "# Cloud Storage bucket and lets you submit training jobs and prediction\n",
         "# requests.\n",
         "\n",
-        "# The Google Cloud Notebook product has specific requirements\n",
-        "IS_GOOGLE_CLOUD_NOTEBOOK = os.path.exists(\"/opt/deeplearning/metadata/env_version\")\n",
+        "import os\n",
+        "import sys\n",
         "\n",
-        "# If on Google Cloud Notebooks, then don't execute this code\n",
-        "if not IS_GOOGLE_CLOUD_NOTEBOOK:\n",
+        "# If on Vertex AI Workbench, then don't execute this code\n",
+        "IS_COLAB = \"google.colab\" in sys.modules\n",
+        "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
+        "    \"DL_ANACONDA_HOME\"\n",
+        "):\n",
         "    if \"google.colab\" in sys.modules:\n",
         "        from google.colab import auth as google_auth\n",
         "\n",
@@ -447,8 +512,8 @@
       },
       "outputs": [],
       "source": [
-        "BUCKET_URI = \"gs://[your-bucket-name]\"  # @param {type:\"string\"}\n",
-        "REGION = \"[your-region]\"  # @param {type:\"string\"}"
+        "BUCKET_NAME = \"[your-bucket-name]\"  # @param {type:\"string\"}\n",
+        "BUCKET_URI = f\"gs://{BUCKET_NAME}\""
       ]
     },
     {
@@ -459,11 +524,9 @@
       },
       "outputs": [],
       "source": [
-        "if BUCKET_URI == \"\" or BUCKET_URI is None or BUCKET_URI == \"gs://[your-bucket-name]\":\n",
-        "    BUCKET_URI = \"gs://\" + PROJECT_ID + \"-aip-\" + TIMESTAMP\n",
-        "\n",
-        "if REGION == \"[your-region]\":\n",
-        "    REGION = \"us-central1\""
+        "if BUCKET_NAME == \"\" or BUCKET_NAME is None or BUCKET_NAME == \"[your-bucket-name]\":\n",
+        "    BUCKET_NAME = PROJECT_ID + \"-aip-\" + TIMESTAMP\n",
+        "    BUCKET_URI = f\"gs://{BUCKET_NAME}\""
       ]
     },
     {
@@ -486,26 +549,6 @@
         "! gsutil mb -l $REGION -p $PROJECT_ID $BUCKET_URI"
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "994afa65eaa2"
-      },
-      "source": [
-        "Run the following cell to grant access to your Cloud Storage resources from Vertex AI Feature store"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "psP1rPU9TRnX"
-      },
-      "outputs": [],
-      "source": [
-        "! gsutil uniformbucketlevelaccess set on $BUCKET_URI"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -526,6 +569,78 @@
         "! gsutil ls -al $BUCKET_URI"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "set_service_account"
+      },
+      "source": [
+        "#### Service Account (Optional)\n",
+        "\n",
+        "If you do not want to use your project's Compute Engine service account, set `SERVICE_ACCOUNT` to another service account ID."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "MQVV9haf2Rra"
+      },
+      "outputs": [],
+      "source": [
+        "SERVICE_ACCOUNT = \"[your-service-account]\"  # @param {type:\"string\"}"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "autoset_service_account"
+      },
+      "outputs": [],
+      "source": [
+        "if (\n",
+        "    SERVICE_ACCOUNT == \"\"\n",
+        "    or SERVICE_ACCOUNT is None\n",
+        "    or SERVICE_ACCOUNT == \"[your-service-account]\"\n",
+        "):\n",
+        "    # Get your service account from gcloud\n",
+        "    if not IS_COLAB:\n",
+        "        shell_output = !gcloud auth list 2>/dev/null\n",
+        "        SERVICE_ACCOUNT = shell_output[2].replace(\"*\", \"\").strip()\n",
+        "\n",
+        "    else:  # IS_COLAB:\n",
+        "        shell_output = ! gcloud projects describe  $PROJECT_ID\n",
+        "        project_number = shell_output[-1].split(\":\")[1].strip().replace(\"'\", \"\")\n",
+        "        SERVICE_ACCOUNT = f\"{project_number}-compute@developer.gserviceaccount.com\"\n",
+        "\n",
+        "    print(\"Service Account:\", SERVICE_ACCOUNT)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "set_service_account:pipelines"
+      },
+      "source": [
+        "#### Set service account access\n",
+        "\n",
+        "Run the following commands to grant your service account access. You only need to run this step once per service account."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "U4UpQThc2Rrb"
+      },
+      "outputs": [],
+      "source": [
+        "! gsutil iam ch serviceAccount:{SERVICE_ACCOUNT}:roles/storage.objectCreator $BUCKET_URI\n",
+        "\n",
+        "! gsutil iam ch serviceAccount:{SERVICE_ACCOUNT}:roles/storage.objectViewer $BUCKET_URI"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -541,13 +656,22 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "G3G2BXqswb_J"
+        "id": "8615339fa4ca"
       },
       "outputs": [],
       "source": [
         "BQ_DATASET = \"Mobile_Gaming\"  # @param {type:\"string\"}\n",
-        "LOCATION = \"US\"\n",
-        "\n",
+        "LOCATION = \"US\""
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "G3G2BXqswb_J"
+      },
+      "outputs": [],
+      "source": [
         "!bq mk --location=$LOCATION --dataset $PROJECT_ID:$BQ_DATASET"
       ]
     },
@@ -600,12 +724,9 @@
       "outputs": [],
       "source": [
         "# Data Engineering and Feature Engineering\n",
-        "TODAY = \"2018-10-03\"\n",
-        "TOMORROW = \"2018-10-04\"\n",
+        "TODAY = \"2022-06-16\"\n",
         "LABEL_TABLE = f\"label_table_{TODAY}\".replace(\"-\", \"\")\n",
-        "FEATURES_TABLE = \"wide_features_table\"  # @param {type:\"string\"}\n",
-        "FEATURES_TABLE_TODAY = f\"wide_features_table_{TODAY}\".replace(\"-\", \"\")\n",
-        "FEATURES_TABLE_TOMORROW = f\"wide_features_table_{TOMORROW}\".replace(\"-\", \"\")\n",
+        "FEATURES_TABLE = f\"wide_features_table_{TODAY}\"  # @param {type:\"string\"}\n",
         "FEATURESTORE_ID = \"mobile_gaming\"  # @param {type:\"string\"}\n",
         "ENTITY_TYPE_ID = \"user\"\n",
         "\n",
@@ -947,13 +1068,37 @@
         "You will cover those steps in details below."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "init_aip:mbsdk,all"
+      },
+      "source": [
+        "### Initialize Vertex AI SDK for Python\n",
+        "\n",
+        "Initialize the Vertex AI SDK for Python for your project and corresponding bucket."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "poLJ0fV52Rrc"
+      },
+      "outputs": [],
+      "source": [
+        "vertex_ai.init(project=PROJECT_ID, location=REGION, staging_bucket=BUCKET_URI)"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "4ffd54e97270"
       },
       "source": [
-        "## Initiate clients"
+        "### Initialize BigQuery SDK for Python\n",
+        "\n",
+        "Initialize the BigQuery AI SDK for Python for your project and corresponding bucket."
       ]
     },
     {
@@ -964,55 +1109,53 @@
       },
       "outputs": [],
       "source": [
-        "bq_client = bigquery.Client(project=PROJECT_ID, location=LOCATION)\n",
-        "vertex_ai.init(project=PROJECT_ID, location=REGION, staging_bucket=BUCKET_URI)"
+        "bq_client = bigquery.Client(project=PROJECT_ID, location=LOCATION)"
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "zmMWIpCwsET9"
+        "id": "WnUQO2IHC9pZ"
       },
       "source": [
         "## Identify users and build your features\n",
-        "\n",
-        "This section we will static features we want to fetch from Vertex AI Feature Store. In particular, we will cover the following steps:\n",
-        "\n",
+        " \n",
+        "This section we will have static features we want to fetch from Vertex AI Feature Store. In particular, we will cover the following steps:\n",
+        " \n",
         "1. Identify users, process demographic features and process behavioral features within the last 24 hours using **BigQuery**\n",
-        "\n",
+        " \n",
         "2. Set up the feature store\n",
-        "\n",
+        " \n",
         "3. Register features using **Vertex AI Feature Store** and the SDK.\n",
-        "\n",
-        "Below you have a picture that shows the process. \n",
-        "\n",
+        " \n",
+        "Below you have a picture that shows the process.\n",
+        " \n",
         "<img src=\"./assets/feature_store_ingestion_2.png\">\n",
-        "\n",
-        "\n",
-        "\n",
-        "The original dataset contains raw event data we cannot ingest in the feature store as they are. We need to pre-process the raw data in order to get user features. \n",
-        "\n",
-        "**Notice we simulate those transformations in different point of time (today and tomorrow).**\n"
+        " \n",
+        " \n",
+        "The original dataset contains raw event data we cannot ingest in the feature store as they are. We need to pre-process the raw data in order to get user features.\n",
+        " \n",
+        "**Notice we simulate those transformations in different points of time (today and tomorrow).**\n"
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "8avYy5QOv02s"
+        "id": "e9zIrwhpDF2q"
       },
       "source": [
         "### Label, Demographic and Behavioral Transformations\n",
-        "\n",
-        "This section is based on the [Churn prediction for game developers using Google Analytics 4 (GA4) and BigQuery ML](https://cloud.google.com/blog/topics/developers-practitioners/churn-prediction-game-developers-using-google-analytics-4-ga4-and-bigquery-ml?utm_source=linkedin&utm_medium=unpaidsoc&utm_campaign=FY21-Q2-Google-Cloud-Tech-Blog&utm_content=google-analytics-4&utm_term=-) blog article by Minhaz Kazi and Polong Lin. \n",
-        "\n",
-        "You will adapt it in order to turn a batch churn prediction (using features within the first 24h user of first engagment) in a real-time churn prediction (using features within the first 24h user of last engagment)."
+        " \n",
+        "This section is based on the [Churn prediction for game developers using Google Analytics 4 (GA4) and BigQuery ML](https://cloud.google.com/blog/topics/developers-practitioners/churn-prediction-game-developers-using-google-analytics-4-ga4-and-bigquery-ml?utm_source=linkedin&utm_medium=unpaidsoc&utm_campaign=FY21-Q2-Google-Cloud-Tech-Blog&utm_content=google-analytics-4&utm_term=-) blog article by Minhaz Kazi and Polong Lin.\n",
+        " \n",
+        "You will adapt it to turn a batch churn prediction (using features within the first 24h user of first engagement) into a real-time churn prediction (using features within the first 6h user of last engagement).\n"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "YO28RAITh6L-"
+        "id": "RQX5m8UiC_px"
       },
       "outputs": [],
       "source": [
@@ -1039,27 +1182,27 @@
         "  SELECT\n",
         "    event_timestamp,\n",
         "    user_pseudo_id,\n",
-        "    SUM(IF(event_name = 'user_engagement', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'user_engagement', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_user_engagement,\n",
-        "    SUM(IF(event_name = 'level_start_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'level_start_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_level_start_quickplay,\n",
-        "    SUM(IF(event_name = 'level_end_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'level_end_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_level_end_quickplay,\n",
-        "    SUM(IF(event_name = 'level_complete_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'level_complete_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_level_complete_quickplay,\n",
-        "    SUM(IF(event_name = 'level_reset_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'level_reset_quickplay', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_level_reset_quickplay,\n",
-        "    SUM(IF(event_name = 'post_score', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'post_score', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_post_score,\n",
-        "    SUM(IF(event_name = 'spend_virtual_currency', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'spend_virtual_currency', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_spend_virtual_currency,\n",
-        "    SUM(IF(event_name = 'ad_reward', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'ad_reward', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_ad_reward,\n",
-        "    SUM(IF(event_name = 'challenge_a_friend', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'challenge_a_friend', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_challenge_a_friend,\n",
-        "    SUM(IF(event_name = 'completed_5_levels', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'completed_5_levels', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_completed_5_levels,\n",
-        "    SUM(IF(event_name = 'use_extra_steps', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 86400000000 PRECEDING\n",
+        "    SUM(IF(event_name = 'use_extra_steps', 1, 0)) OVER (PARTITION BY user_pseudo_id ORDER BY event_timestamp ASC RANGE BETWEEN 21600000000 PRECEDING\n",
         "      AND CURRENT ROW ) AS cnt_use_extra_steps,\n",
         "  FROM (\n",
         "    SELECT\n",
@@ -1071,7 +1214,7 @@
         "\n",
         "SELECT\n",
         "    -- PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', CONCAT('{TODAY}', ' ', STRING(TIME_TRUNC(CURRENT_TIME(), SECOND))), 'UTC') as timestamp,\n",
-        "    PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', FORMAT_TIMESTAMP('%Y-%m-%d %H:%M:%S', TIMESTAMP_MICROS(beh.event_timestamp))) AS timestamp,\n",
+        "    TIMESTAMP_ADD(PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', FORMAT_TIMESTAMP('%Y-%m-%d %H:%M:%S', TIMESTAMP_MICROS(beh.event_timestamp))), INTERVAL 1351 DAY) AS timestamp,\n",
         "    dem.*,\n",
         "    CAST(IFNULL(beh.cnt_user_engagement, 0) AS FLOAT64)  AS cnt_user_engagement,\n",
         "    CAST(IFNULL(beh.cnt_level_start_quickplay, 0) AS FLOAT64) AS cnt_level_start_quickplay,\n",
@@ -1097,7 +1240,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "Z6CjIOmDsET-"
+        "id": "oGYxLCSnD068"
       },
       "outputs": [],
       "source": [
@@ -1107,31 +1250,31 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "Lx__2-assET-"
+        "id": "xQLIlsTCD_nk"
       },
       "source": [
         "## Create a Vertex AI Feature store and ingest your features\n",
-        "\n",
-        "Now you have the wide table of features. It is time to ingest them into the feature store. \n",
-        "\n",
+        " \n",
+        "Now you have a wide table of features. It is time to ingest them into the feature store.\n",
+        " \n",
         "Before to moving on, you may have a question: **Why do I need a feature store**\n",
         "in this scenario at that point?\n",
-        "\n",
-        "One of the reason would be to make those features accessable across team by calculating once and reuse them many times. And in order to make it possible you need also be able to monitor those features over time to guarantee freshness and in case have a new feature engineerign run to refresh them. \n",
-        "\n",
-        "If it is not your case, I will give even more reasons about why you should consider feature store in the following sections. Just keep following me for now.\n",
-        "\n",
-        "One of the most important thing is related to its data model. As you can see in the picture below, Vertex AI Feature Store organizes resources hierarchically in the following order: `Featurestore -> EntityType -> Feature`. You must create these resources before you can ingest data into Vertex AI Feature Store.\n",
-        "\n",
+        " \n",
+        "One of the reasons would be to make those features accessible across teams by calculating once and reuse them many times. And in order to make it possible you need also be able to monitor those features over time to guarantee freshness and in case have a new feature engineering run to refresh them.\n",
+        " \n",
+        "If it is not your case, I will give even more reasons about why you should consider a feature store in the following sections. Just keep following me for now.\n",
+        " \n",
+        "One of the most important things is related to its data model. As you can see in the picture below, Vertex AI Feature Store organizes resources hierarchically in the following order: `Featurestore -> EntityType -> Feature`. You must create these resources before you can ingest data into Vertex AI Feature Store.\n",
+        " \n",
         "<img src=\"./assets/feature_store_data_model_3.png\">\n",
-        "\n",
-        "In our case we are going to create **mobile_gaming** featurestore resource containing **user** entity type and all its associated **features** such as country or the number of times a user challenged a friend (cnt_challenge_a_friend)."
+        " \n",
+        "In our case we are going to create **mobile_gaming** featurestore resource containing **user** entity type and all its associated **features** such as country or the number of times a user challenged a friend (cnt_challenge_a_friend).\n"
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "8dNlxda2sET_"
+        "id": "VR7BJEozED_Q"
       },
       "source": [
         "### Create featurestore, ```mobile_gaming```\n",
@@ -1143,7 +1286,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "t2en8I7TSe4b"
+        "id": "vUFqtYU-EDTR"
       },
       "outputs": [],
       "source": [
@@ -1164,7 +1307,7 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "rN-vlvPUsET_"
+        "id": "mUlCwfdpEHJG"
       },
       "source": [
         "### Create the ```User``` entity type and its features\n",
@@ -1176,7 +1319,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "CbZ2RQ5XbuRq"
+        "id": "PnCU1wBND3W7"
       },
       "outputs": [],
       "source": [
@@ -1194,7 +1337,7 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "B2PIAprPmnhB"
+        "id": "bT9LXzu1EOvW"
       },
       "source": [
         "### Set Feature Monitoring\n",
@@ -1208,7 +1351,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "N6im2c3ymiwC"
+        "id": "8WBlYUkOERaI"
       },
       "outputs": [],
       "source": [
@@ -1231,7 +1374,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "gp9xaLQXn0CS"
+        "id": "92X4-7PFETj5"
       },
       "outputs": [],
       "source": [
@@ -1254,18 +1397,18 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "ustwKOMle8Qp"
+        "id": "hxAuZjt3EWFo"
       },
       "source": [
         "### Create features\n",
         "\n",
-        "In order to ingest features, you need to provide feature configuration and create them as featurestore resources.\n"
+        "In order to ingest features, you need to provide feature configuration and create them as featurestore resources."
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "ijeZCTKIfCRL"
+        "id": "hRXO2I5VEYwt"
       },
       "source": [
         "#### Create Feature configuration\n",
@@ -1278,7 +1421,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "vX_uYmjUgd9x"
+        "id": "K26NEYZIEbvE"
       },
       "outputs": [],
       "source": [
@@ -1359,7 +1502,7 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "ErkruXPJkPuy"
+        "id": "FjzMd1XbEfdo"
       },
       "source": [
         "#### Create features using `batch_create_features` method\n",
@@ -1371,7 +1514,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "ZsCAO_IfsEUC"
+        "id": "nqlgCDI9pbCD"
       },
       "outputs": [],
       "source": [
@@ -1388,19 +1531,19 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "9WisJk18qqgs"
+        "id": "7zpFV7wAppkC"
       },
       "source": [
         "### Search features\n",
         "\n",
-        "Vertex AI Feature store supports serching capabilities. Below you have a simple example that show how to filter a feature based on its name. "
+        "Vertex AI Feature store supports searching capabilities. Below you have a simple example that shows how to filter a feature based on its name. "
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "JzqyarMZqvZS"
+        "id": "BJXYLLOfppCL"
       },
       "outputs": [],
       "source": [
@@ -1412,7 +1555,7 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "ugtBfW5gsEUD"
+        "id": "is9C_6-QpxG3"
       },
       "source": [
         "## Ingest features \n",
@@ -1447,7 +1590,7 @@
         "        entity_id_field=ENTITY_ID_FIELD,\n",
         "        disable_online_serving=False,\n",
         "        worker_count=10,\n",
-        "        sync=True,\n",
+        "        sync=False,\n",
         "    )\n",
         "except RuntimeError as error:\n",
         "    print(error)"
@@ -1456,7 +1599,7 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "3Yv8MenWXrRX"
+        "id": "8lCMpDPGp-oQ"
       },
       "source": [
         "# Train and deploy a real-time churn ML model using Vertex AI Training and Endpoints\n",
@@ -1467,34 +1610,34 @@
         "\n",
         "<img src=\"./assets/train_model_4.png\">\n",
         "\n",
-        "Let's dive into each step of this process.\n"
+        "Let's dive into each step of this process."
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "saZZ3zWKX1YK"
+        "id": "VMrvnuyjqGfY"
       },
       "source": [
-        "## Fetch training data with point-in-time query using BigQuery and Vertex AI Feature store  \n",
-        "\n",
-        "As we mentioned above, in real time churn prediction, it is so important defining the label you want to predict with your model. \n",
-        "\n",
-        "Let's assume that you decide to predict the churn probability over the last 24 hr. So now you have your label. Next step is to define your training sample. But let's think about that for a second. \n",
-        "\n",
-        "In that churn real time system, you have a high volume of transactions you could use to calculate those features which keep floating and are collected constantly over time. It implies that you always get fresh data to reconstruct features. And depending on when you decide to calculate one feature or another you can end up with a set of features that are not aligned in time. \n",
-        "\n",
+        "## Fetch training data with point-in-time query using BigQuery and Vertex AI Feature store \n",
+        " \n",
+        "As we mentioned above, in real time churn prediction, it is so important defining the label you want to predict with your model.\n",
+        " \n",
+        "Let's assume that you decide to predict the churn probability over the next hour. So now you have your label. Next step is to define your training sample. But let's think about that for a second.\n",
+        " \n",
+        "In that churn real time system, you have a high volume of transactions you could use to calculate those features which keep floating and are collected constantly over time. It implies that you always get fresh data to reconstruct features. And depending on when you decide to calculate one feature or another you can end up with a set of features that are not aligned in time.\n",
+        " \n",
         "When you have labels available, it would be incredibly difficult to say which set of features contains the most up to date historical information associated with the label you want to predict. And, when you are not able to guarantee that, the performance of your model would be badly affected because you serve no representative features of the data and the label from the field when it goes live. So you need a way to get the most updated features you calculated over time before the label becomes available in order to avoid this informational skew.\n",
-        "\n",
-        "**With the Vertex AI Feature store, you can fetch feature values corresponding to a particular timestamp thanks to point-in-time lookup capability.** In our case, it would be the timestamp associated to the label you want to predict with your model. In this way, you will avoid data leakage and you will get the most updated features to train your model. \n",
-        "\n",
-        "Let's see how to do that. \n"
+        " \n",
+        "**With the Vertex AI Feature store, you can fetch feature values corresponding to a particular timestamp thanks to point-in-time lookup capability.** In our case, it would be the timestamp associated with the label you want to predict with your model. In this way, you will avoid data leakage and you will get the most updated features to train your model.\n",
+        " \n",
+        "Let's see how to do that.\n"
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "YHNbIHqFcQiM"
+        "id": "RE_Pvmu-qdDt"
       },
       "source": [
         "### Define query for reading instances at a specific point in time\n",
@@ -1506,7 +1649,7 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "DGUm0bYqhVV4"
+        "id": "bUDVw7l-qF2x"
       },
       "outputs": [],
       "source": [
@@ -1518,13 +1661,13 @@
         "  # get training threshold ----------------------------------------------------------------------------------\n",
         "  get_training_threshold AS (\n",
         "  SELECT\n",
-        "    (MAX(event_timestamp) - 86400000000) AS training_thrs\n",
+        "    (MAX(event_timestamp) - 10800000000) AS training_thrs\n",
         "  FROM\n",
         "    `firebase-public-project.analytics_153293282.events_*`\n",
         "  WHERE\n",
         "    event_name=\"user_engagement\"\n",
         "    AND\n",
-        "    PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', FORMAT_TIMESTAMP('%Y-%m-%d %H:%M:%S', TIMESTAMP_MICROS(event_timestamp))) < '{TODAY}'),\n",
+        "    TIMESTAMP_ADD(PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', FORMAT_TIMESTAMP('%Y-%m-%d %H:%M:%S', TIMESTAMP_MICROS(event_timestamp))), INTERVAL 1351 DAY) < '{TODAY}'),\n",
         "\n",
         "  # query to create label -----------------------------------------------------------------------------------\n",
         "  get_label AS (\n",
@@ -1549,7 +1692,7 @@
         "    WHERE\n",
         "      event_name=\"user_engagement\"\n",
         "    AND\n",
-        "    PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', FORMAT_TIMESTAMP('%Y-%m-%d %H:%M:%S', TIMESTAMP_MICROS(event_timestamp))) < '{TODAY}'\n",
+        "    TIMESTAMP_ADD(PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', FORMAT_TIMESTAMP('%Y-%m-%d %H:%M:%S', TIMESTAMP_MICROS(event_timestamp))), INTERVAL 1351 DAY) < '{TODAY}'\n",
         "    GROUP BY\n",
         "      user_pseudo_id )\n",
         "  GROUP BY\n",
@@ -1669,7 +1812,7 @@
       "source": [
         "!mkdir -m 777 -p trainer data/ingest data/raw model config\n",
         "!gsutil -m cp -r $GCS_DESTINATION_OUTPUT_URI/*.csv data/ingest\n",
-        "!head -n 1000 data/ingest/*.csv > data/raw/sample.csv"
+        "!head -n 2000 data/ingest/*.csv > data/raw/sample.csv"
       ]
     },
     {
@@ -2071,7 +2214,7 @@
       },
       "outputs": [],
       "source": [
-        "TRAIN_JOB_RESOURCE_NAME = \"\"  # @param {type:\"string\"}"
+        "TRAIN_JOB_RESOURCE_NAME = \"[your-train-job-resource-name]\"  # @param {type:\"string\"}"
       ]
     },
     {
@@ -2166,32 +2309,32 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "7c9330928aa1"
+        "id": "1TNzL_EGrVUm"
       },
       "source": [
         "# Serve ML features at scale with low latency\n",
-        "\n",
-        "At that time, you are ready **to deploy our simple model which would requires fetching preprocessed attributes as input features in real time**. \n",
-        "\n",
+        " \n",
+        "At that time, you are ready **to deploy our simple model which would requires fetching preprocessed attributes as input features in real time**.\n",
+        " \n",
         "Below you can see how it works\n",
-        "\n",
-        "<img src=\"./assets/online_serving_5.png\" width=\"600\">\n",
-        "\n",
-        "But think about those features for a second. \n",
-        "\n",
-        "Your behavioral features used to trained your model, they cannot be computed when you are going to serve the model online. \n",
-        "\n",
-        "How could you compute the number of time a user challenged a friend withing the last 24 hours on the fly?\n",
-        "\n",
-        "You simply can't do that. You need to be computed this feature on the server side and serve it with low latency. And becuase Bigquery is not optimized for those read operations, we need a different service that allows singleton lookup where the result is a single row with many columns.\n",
-        "\n",
-        "Also, even if it was not the case, when you deploy a model that requires preprocessing your data, you need to be sure to reproduce the same preprocessing steps you had when you trained it. If you are not able to do that a skew between training and serving data would happen and it will affect badly your model performance (and in the worst scenario break your serving system). \n",
-        "\n",
-        "You need a way to mitigate that in a way you don't need to implement those preprocessing steps online but just serve the same aggregated features you already have for training to generate online prediction. \n",
-        "\n",
-        "These are other valuable reasons to introduce Vertex AI Feature Store. With it, you have a service which helps you to serve feature at scale with low latency as they were available at training time mitigating in that way possible training-serving skew.\n",
-        "\n",
-        "Now that you know **why you need a feature store**, let's closing this journey by deploying your model and use feature store to retrieve features online, pass them to endpoint and generate predictions.\n"
+        " \n",
+        "<center><img src=\"./assets/online_serving_5.png\" width=\"800\"/></center>\n",
+        " \n",
+        "But think about those features for a second.\n",
+        " \n",
+        "Your behavioral features used to train your model, they cannot be computed when you are going to serve the model online.\n",
+        " \n",
+        "How could you compute the number of times a user challenged a friend within the last 24 hours on the fly?\n",
+        " \n",
+        "You need to be computed this feature on the server side and serve it with low latency. And because Bigquery is not optimized for those read operations, we need a different service that allows singleton lookup where the result is a single row with many columns.\n",
+        " \n",
+        "Also, even if it was not the case, when you deploy a model that requires preprocessing your data, you need to be sure to reproduce the same preprocessing steps you had when you trained it. If you are not able to do that a skew between training and serving data would happen and it will badly affect your model performance (and in the worst scenario break your serving system).\n",
+        " \n",
+        "You need a way to mitigate that in a way you don't need to implement those preprocessing steps online but just serve the same aggregated features you already have for training to generate online prediction.\n",
+        " \n",
+        "These are other valuable reasons to introduce Vertex AI Feature Store. With it, you have a service which helps you to serve features at scale with low latency as they were available at training time mitigating in that way possible training-serving skew.\n",
+        " \n",
+        "Now that you know **why you need a feature store**, let's conclude this journey by deploying your model using a feature store to retrieve features online, pass them to the endpoint and generate predictions.\n"
       ]
     },
     {
@@ -2221,13 +2364,13 @@
       },
       "outputs": [],
       "source": [
-        "simulate_prediction(endpoint=endpoint, n_requests=1000, latency=1)"
+        "simulate_prediction(endpoint=endpoint, n_requests=10, latency=1)"
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "TpV-iwP9qw9c"
+        "id": "8d3S1d1urZOy"
       },
       "source": [
         "## Cleaning up\n",
@@ -2267,11 +2410,12 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "sx_vKniMq9ZX"
+        "id": "FMXT2akXrZOy"
       },
       "outputs": [],
       "source": [
         "# Delete bucket\n",
+        "delete_bucket = False\n",
         "if (delete_bucket or os.getenv(\"IS_TESTING\")) and \"BUCKET_URI\" in globals():\n",
         "    ! gsutil -m rm -r $BUCKET_URI"
       ]

notebooks/community/ml_ops/stage1/README.md

@@ -28,11 +28,50 @@ The first stage in MLOps is the collection and preparation for the purpose of de
 
 ### Get Started
 
-[Get Started with BQ datasets](get_started_bq_datasets.ipynb)
+[Get started with Vertex AI datasets](get_started_vertex_datasets.ipynb)
 
 ```
 The steps performed include:
+- Create a Vertex AI `Dataset` resource for:
+    - image data
+    - text data
+    - video data
+    - tabular data
+    - forecasting data
+- Search `Dataset` resources using a filter.
+- Read a sample of a `BigQuery` dataset into a dataframe.
+- Generate statistics and data schema using TensorFlow Data Validation from the samples in the dataframe.
+- Detect anomalies in new data using TensorFlow Data Validation.
+- Generate a TFRecord feature specification using TensorFlow Transform from the data schema.
+- Export a dataset and convert to TFRecords.
+```
 
+[Get started with Dataflow](get_started_dataflow.ipynb)
+
+```
+The steps performed include:
+- Offline preprocessing of data:
+    - Serially - w/o dataflow
+    - Parallel - with dataflow
+- Upstream preprocessing of data:
+    - tabular data
+    - image data
+```
+
+[Create an unlabelled Vertex AI AutoML text entity extraction dataset from pdfs using Vision API](get_started_with_visionapi_and_vertex_datasets.ipynb)
+
+```
+The steps performed include:
+1. Using `Vision API` to perform Optical Character Recognition (OCR) to extract text from PDF files.
+2. Processing the results and saving them to text files.
+3. Generating a `Vertex AI Dataset` import file.
+4. Creating a new unlabelled text entity extraction `Vertex AI Dataset` resource in `Vertex AI`.
+```
+
+[Get started with BigQuery datasets](get_started_bq_datasets.ipynb)
+
+```
+The steps performed include:
 - Create a Vertex AI `Dataset` resource from `BigQuery` table -- compatible for `AutoML` training.
 - Extract a copy of the dataset from `BigQuery` to a CSV file in Cloud Storage -- compatible for `AutoML` or custom training.
 - Select rows from a `BigQuery` dataset into a `pandas` dataframe -- compatible for custom training.
@@ -42,59 +81,25 @@ The steps performed include:
 - Extract data from `BigQuery` table into a `DMatrix` -- compatible for custom training `XGBoost` models.
 ```
 
-[Get Started with Vertex datasets](get_started_vertex_datasets.ipynb)
+[Get started with Vertex AI data labeling](get_started_with_data_labeling.ipynb)
 
 ```
 The steps performed include:
-  
-- Create a Vertex AI `Dataset` resource for:
-    - image data
-    - text data
-    - video data
-    - tabular data
-    - forecasting data
-
-
-- Search `Dataset` resources using a filter.
-- Read a sample of a `BigQuery` dataset into a dataframe.
-- Generate statistics and data schema using TensorFlow Data Validation from the samples in the dataframe.
-- Detect anomalies in new data using TensorFlow Data Validation.
-- Generate a TFRecord feature specification using TensorFlow Transform from the data schema.
-- Export a dataset and convert to TFRecords.
-```
-
-[Get Started with Dataflow](get_started_dataflow.ipynb)
-
-```
-The steps performed include:
-
-- Offline preprocessing of data:
-    - Serially - w/o dataflow
-    - Parallel - with dataflow
-- Upstream preprocessing of data:
-    - tabular data
-    - image data
-```
-
-[Get Started with Data Labeling](get_started_data_labeling.ipynb)
-
-```
-The steps performed include:
-
 - Create a Specialist Pool for data labelers.
 - Create a data labeling job.
 - Submit the data labeling job.
 - List data labeling jobs.
 - Cancel a data labeling job.
+
 ```
 
 ### E2E Stage Example
 
 [Stage 1: Data Management](mlops_data_management.ipynb)
-  
+
+
 ```
 The steps performed include:
-
 - Explore and visualize the data.
 - Create a Vertex AI `Dataset` resource from `BigQuery` table -- for AutoML training.
 - Extract a copy of the dataset to a CSV file in Cloud Storage.
@@ -104,4 +109,3 @@ The steps performed include:
 - Generate a TFRecord feature specification using TensorFlow Data Validation from the data schema.
 - Preprocess a portion of the BigQuery data using `Dataflow` -- for custom training.
 ```
-

notebooks/community/ml_ops/stage1/get_started_bq_datasets.ipynb

@@ -64,17 +64,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 1 : data management: get started with BigQuery datasets."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:gsod,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). In this version of the dataset you consider the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -140,8 +129,26 @@
         "- Alternatively:\n",
         "    - Extract the BigQuery table to CSV files.\n",
         "    - Preprocess the CSV files.\n",
-        "    - Create a tf.data.Dataset generator from the CSV files.\n",
-        "    \n",
+        "    - Create a tf.data.Dataset generator from the CSV files."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:gsod,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). In this version of the dataset you consider the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "9e483012a752"
+      },
+      "source": [
         "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -174,7 +181,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -389,7 +396,7 @@
       },
       "outputs": [],
       "source": [
-        "i  # If you are running this notebook in Colab, run this cell and follow the\n",
+        "# If you are running this notebook in Colab, run this cell and follow the\n",
         "# instructions to authenticate your GCP account. This provides access to your\n",
         "# Cloud Storage bucket and lets you submit training jobs and prediction\n",
         "# requests.\n",
@@ -1132,7 +1139,7 @@
       "source": [
         "dataframe[\"station_number\"] = pd.to_numeric(dataframe[\"station_number\"])\n",
         "labels = dataframe[\"mean_temp\"]\n",
-        "data = dataframe.drop(4)\n",
+        "data = dataframe.drop([\"mean_temp\"], axis=1)\n",
         "\n",
         "dtrain = xgb.DMatrix(data, label=labels)"
       ]

notebooks/community/ml_ops/stage1/get_started_dataflow.ipynb

@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 1 : data management: get started with Dataflow."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:gsod,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -137,6 +126,34 @@
         "Alternately for AutoML tabular model training, you can reconfigure the otherwise default preprocessing."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:gsod,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "9e483012a752"
+      },
+      "source": [
+        "### Costs\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "- Vertex AI\n",
+        "- Cloud Storage\n",
+        "- BigQuery\n",
+        "- Dataflow\n",
+        "\n",
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing), [BigQuery pricing](https://cloud.google.com/bigquery/pricing), and [Dataflow pricing](https://cloud.google.com/dataflow/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -159,7 +176,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",

notebooks/community/ml_ops/stage1/get_started_vertex_datasets.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 1 : data management: get started with Vertex datasets\n",
+        "# E2E ML on GCP: MLOps stage 1 : data management: get started with Vertex AI datasets\n",
         "\n",
         "<table align=\"left\">\n",
         "  <td>\n",
@@ -136,9 +136,26 @@
         "            - Create a tf.data.Dataset generator from the CSV index file.\n",
         "        - If text strings are in text files:\n",
         "            - Using the JSON index file, convert the text files and labels to TFRecords.\n",
-        "            - Create a tf.data.Dataset from the TFRecords.\n",
+        "            - Create a tf.data.Dataset from the TFRecords."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "533dd6fe83c8"
+      },
+      "source": [
+        "### Datasets\n",
         "\n",
-        "    \n",
+        "This tutorial uses a variety of public datasets to demonstrate using a `Vertex AI` managed dataset."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "9e483012a752"
+      },
+      "source": [
         "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -171,7 +188,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -188,7 +205,7 @@
         "! pip3 install --upgrade google-cloud-aiplatform $USER_FLAG -q\n",
         "! pip3 install --upgrade google-cloud-bigquery $USER_FLAG -q\n",
         "! pip3 install -U tensorflow-io==0.18 $USER_FLAG -q\n",
-        "! pip3 install --upgrade future $USER_FLAG -q"
+        "! pip3 install --upgrade db-dtypes $USER_FLAG -q! pip3 install --upgrade future $USER_FLAG -q"
       ]
     },
     {
@@ -226,6 +243,8 @@
         "id": "cb082379ed5b"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -263,36 +282,22 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
-        "PROJECT_ID = \"\"\n",
-        "\n",
-        "if not os.getenv(\"IS_TESTING\"):\n",
-        "    # Get your Google Cloud project ID from gcloud\n",
-        "    shell_output = !gcloud config list --format 'value(core.project)' 2>/dev/null\n",
-        "    PROJECT_ID = shell_output[0]\n",
-        "    print(\"Project ID: \", PROJECT_ID)"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "37c0a68ff20d"
-      },
-      "source": [
-        "Otherwise, set your project ID here."
+        "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "autoset_project_id"
+        "id": "nWlzLu5ELxWd"
       },
       "outputs": [],
       "source": [
-        "if PROJECT_ID == \"\" or PROJECT_ID is None:\n",
-        "    PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
+        "if PROJECT_ID == \"\" or PROJECT_ID is None or PROJECT_ID == \"[your-project-id]\":\n",
+        "    # Get your GCP project id from gcloud\n",
+        "    shell_output = ! gcloud config list --format 'value(core.project)' 2>/dev/null\n",
+        "    PROJECT_ID = shell_output[0]\n",
+        "    print(\"Project ID:\", PROJECT_ID)"
       ]
     },
     {
@@ -408,12 +413,11 @@
         "import sys\n",
         "\n",
         "# If on Vertex AI Workbench, then don't execute this code\n",
-        "IS_COLAB = False\n",
+        "IS_COLAB = \"google.colab\" in sys.modules\n",
         "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
         "    \"DL_ANACONDA_HOME\"\n",
         "):\n",
         "    if \"google.colab\" in sys.modules:\n",
-        "        IS_COLAB = True\n",
         "        from google.colab import auth as google_auth\n",
         "\n",
         "        google_auth.authenticate_user()\n",
@@ -558,7 +562,7 @@
       },
       "outputs": [],
       "source": [
-        "aip.init(project=PROJECT_ID, location=REGION)"
+        "aip.init(project=PROJECT_ID, location=REGION, staging_bucket=BUCKET_URI)"
       ]
     },
     {
@@ -612,26 +616,13 @@
         "Learn more about [All dataset documentation](https://cloud.google.com/vertex-ai/docs/datasets/datasets)"
       ]
     },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "import_file:flowers,csv,icn"
-      },
-      "outputs": [],
-      "source": [
-        "IMPORT_FILE = (\n",
-        "    \"gs://cloud-samples-data/vision/automl_classification/flowers/all_data_v2.csv\"\n",
-        ")"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "create_dataset:image,icn"
       },
       "source": [
-        "### Create the Dataset\n",
+        "### Create an Image Dataset\n",
         "\n",
         "Next, create the `Dataset` resource using the `create` method for the `ImageDataset` class, which takes the following parameters:\n",
         "\n",
@@ -646,6 +637,19 @@
         "Learn more about [ImageDataset](https://cloud.google.com/vertex-ai/docs/datasets/prepare-image)."
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "import_file:flowers,csv,icn"
+      },
+      "outputs": [],
+      "source": [
+        "IMPORT_FILE = (\n",
+        "    \"gs://cloud-samples-data/vision/automl_classification/flowers/all_data_v2.csv\"\n",
+        ")"
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,
@@ -663,24 +667,13 @@
         "print(dataset.resource_name)"
       ]
     },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "import_file:hmdb,csv,vcn"
-      },
-      "outputs": [],
-      "source": [
-        "IMPORT_FILE = \"gs://automl-video-demo-data/hmdb_split1_5classes_train_inf.csv\""
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "create_dataset:video,vcn"
       },
       "source": [
-        "### Create the Dataset\n",
+        "### Create a Video Dataset\n",
         "\n",
         "Next, create the `Dataset` resource using the `create` method for the `VideoDataset` class, which takes the following parameters:\n",
         "\n",
@@ -694,6 +687,17 @@
         "Learn more about [VideoDataset](https://cloud.google.com/vertex-ai/docs/datasets/prepare-video)."
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "import_file:hmdb,csv,vcn"
+      },
+      "outputs": [],
+      "source": [
+        "IMPORT_FILE = \"gs://automl-video-demo-data/hmdb_split1_5classes_train_inf.csv\""
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,
@@ -711,24 +715,13 @@
         "print(dataset.resource_name)"
       ]
     },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "import_file:happydb,csv,tcn"
-      },
-      "outputs": [],
-      "source": [
-        "IMPORT_FILE = \"gs://cloud-ml-data/NL-classification/happiness.csv\""
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "create_dataset:text,tcn"
       },
       "source": [
-        "### Create the Dataset\n",
+        "### Create a Text Dataset\n",
         "\n",
         "Next, create the `Dataset` resource using the `create` method for the `TextDataset` class, which takes the following parameters:\n",
         "\n",
@@ -743,6 +736,17 @@
         "Learn more about [TextDataset](https://cloud.google.com/vertex-ai/docs/datasets/prepare-text)."
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "import_file:happydb,csv,tcn"
+      },
+      "outputs": [],
+      "source": [
+        "IMPORT_FILE = \"gs://cloud-ml-data/NL-classification/happiness.csv\""
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,
@@ -760,6 +764,24 @@
         "print(dataset.resource_name)"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "create_dataset:tabular,bq,lrg,v2"
+      },
+      "source": [
+        "### Create a Tabular Dataset\n",
+        "\n",
+        "#### CSV input data\n",
+        "\n",
+        "Next, create the `Dataset` resource using the `create` method for the `TabularDataset` class for CSV input data, which takes the following parameters:\n",
+        "\n",
+        "- `display_name`: The human readable name for the `Dataset` resource.\n",
+        "- `gcs_source`: A list of one or more dataset index files to import the data items into the `Dataset` resource.\n",
+        "\n",
+        "Learn more about [TabularDataset from CSV files](https://cloud.google.com/vertex-ai/docs/datasets/create-dataset-api#aiplatform_create_dataset_tabular_gcs_sample-python)"
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,
@@ -768,27 +790,50 @@
       },
       "outputs": [],
       "source": [
-        "IMPORT_FILE = \"bq://bigquery-public-data.samples.gsod\"\n",
-        "BQ_TABLE = \"bigquery-public-data.samples.gsod\""
+        "IMPORT_FILE = \"gs://cloud-samples-data/tables/iris_1000.csv\""
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "create_dataset:tabular,bq,lrg,v2"
+      },
+      "outputs": [],
+      "source": [
+        "dataset = aip.TabularDataset.create(\n",
+        "    display_name=\"example\" + \"_\" + TIMESTAMP, gcs_source=[IMPORT_FILE]\n",
+        ")\n",
+        "\n",
+        "print(dataset.resource_name)"
       ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "create_dataset:tabular,bq,lrg,v2"
+        "id": "854dd1e0195c"
       },
       "source": [
-        "### Create the Dataset\n",
+        "#### BigQuery input data\n",
         "\n",
-        "#### CSV input data\n",
-        "\n",
-        "Next, create the `Dataset` resource using the `create` method for the `TabularDataset` class, which takes the following parameters:\n",
+        "Next, create the `Dataset` resource using the `create` method for the `TabularDataset` class for BigQuery table input, which takes the following parameters:\n",
         "\n",
         "- `display_name`: The human readable name for the `Dataset` resource.\n",
-        "- `gcs_source`: A list of one or more dataset index files to import the data items into the `Dataset` resource.\n",
-        "- `labels`: User defined metadata. In this example, you store the location of the Cloud Storage bucket containing the user defined data.\n",
+        "- `bq_source`: A list of one or more BigQuery tables to import the data items into the `Dataset` resource.\n",
         "\n",
-        "Learn more about [TabularDataset from CSV files](https://cloud.google.com/vertex-ai/docs/datasets/create-dataset-api#aiplatform_create_dataset_tabular_gcs_sample-python)"
+        "Learn more about [TabularDataset from BigQuery table](https://cloud.google.com/vertex-ai/docs/datasets/create-dataset-api#aiplatform_create_dataset_tabular_bigquery_sample-pythonn)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "86343c146300"
+      },
+      "outputs": [],
+      "source": [
+        "IMPORT_FILE = \"bq://bigquery-public-data.samples.gsod\"\n",
+        "BQ_TABLE = \"bigquery-public-data.samples.gsod\""
       ]
     },
     {
@@ -806,15 +851,63 @@
         "print(dataset.resource_name)"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "82e9fe20ce71"
+      },
+      "source": [
+        "#### Dataframe input data\n",
+        "\n",
+        "Next, create the `Dataset` resource using the `create_from_dataframe` method for the `TabularDataset` class for pandas dataframe input, which takes the following parameters:\n",
+        "\n",
+        "- `display_name`: The human readable name for the `Dataset` resource.\n",
+        "- `df_source`: The pandas dataframe to import the data items into the `Dataset` resource.\n",
+        "- `staging_path`: The BigQuery table to store the imported data."
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "import_file:covid,csv,forecast"
+        "id": "3805f945ffdd"
       },
       "outputs": [],
       "source": [
-        "IMPORT_FILE = \"gs://cloud-samples-data/ai-platform/covid/bigquery-public-covid-nyt-us-counties-train.csv\""
+        "# Download the table.\n",
+        "table = bigquery.TableReference.from_string(BQ_TABLE)\n",
+        "\n",
+        "rows = bqclient.list_rows(\n",
+        "    table,\n",
+        "    max_results=10000,\n",
+        "    selected_fields=[\n",
+        "        bigquery.SchemaField(\"station_number\", \"STRING\"),\n",
+        "        bigquery.SchemaField(\"year\", \"INTEGER\"),\n",
+        "        bigquery.SchemaField(\"month\", \"INTEGER\"),\n",
+        "        bigquery.SchemaField(\"day\", \"INTEGER\"),\n",
+        "        bigquery.SchemaField(\"mean_temp\", \"FLOAT\"),\n",
+        "    ],\n",
+        ")\n",
+        "\n",
+        "dataframe = rows.to_dataframe()\n",
+        "print(dataframe.head())"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "create_dataset:tabular,bq,lrg,v2"
+      },
+      "outputs": [],
+      "source": [
+        "dataset = aip.TabularDataset.create_from_dataframe(\n",
+        "    display_name=\"example\" + \"_\" + TIMESTAMP,\n",
+        "    df_source=dataframe,\n",
+        "    staging_path=f\"bq://{PROJECT_ID}.samples.gsod\",\n",
+        ")\n",
+        "\n",
+        "print(dataset.resource_name)"
       ]
     },
     {
@@ -823,7 +916,7 @@
         "id": "create_dataset:tabular,forecast,v2"
       },
       "source": [
-        "### Create the Dataset\n",
+        "### Create a Time Series Dataset\n",
         "\n",
         "Next, create the `Dataset` resource using the `create` method for the `TimeSeriesDataset` class, which takes the following parameters:\n",
         "\n",
@@ -834,6 +927,17 @@
         "Learn more about [TimeSeriesDataset](https://cloud.google.com/vertex-ai/docs/datasets/prepare-tabular)."
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "import_file:covid,csv,forecast"
+      },
+      "outputs": [],
+      "source": [
+        "IMPORT_FILE = \"gs://cloud-samples-data/ai-platform/covid/bigquery-public-covid-nyt-us-counties-train.csv\""
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,
@@ -1279,7 +1383,7 @@
         "import os\n",
         "import sys\n",
         "\n",
-        "# If on Google Cloud Notebook, then don't execute this code\n",
+        "# If on Workbench AI Notebook, then don't execute this code\n",
         "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\"):\n",
         "    if \"google.colab\" in sys.modules:\n",
         "        ! pip3 install fsspec\n",
@@ -1620,11 +1724,15 @@
       },
       "outputs": [],
       "source": [
+        "delete_bucket = False\n",
+        "\n",
         "# Delete the dataset using the Vertex dataset object\n",
-        "dataset.delete()\n",
+        "datasets = aip.TabularDataset.list(filter=f'display_name=\"example_{TIMESTAMP}\"')\n",
+        "for dataset in datasets:\n",
+        "    dataset.delete()\n",
         "\n",
         "# Delete the bucket\n",
-        "if os.getenv(\"IS_TESTING\"):\n",
+        "if delete_bucket or os.getenv(\"IS_TESTING\"):\n",
         "    ! gsutil rm -r $BUCKET_URI"
       ]
     }

notebooks/community/ml_ops/stage1/get_started_with_data_labeling.ipynb

@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 1 : data management: get started with Vertex AI Data Labeling service."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -102,6 +91,17 @@
         "Learn more about [Request a Vertex AI Data Labeling job](https://cloud.google.com/vertex-ai/docs/datasets/data-labeling-job)."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -146,7 +146,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -167,7 +167,7 @@
         "id": "restart"
       },
       "source": [
-        "### Restart the Kernel\n",
+        "### Restart the kernel\n",
         "\n",
         "Once you've installed the Vertex AI SDK and Google *cloud-storage*, you need to restart the notebook kernel so it can find the packages.\n"
       ]
@@ -220,6 +220,17 @@
         "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands.\n"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "WReHDGG5g0XY"
+      },
+      "source": [
+        "#### Set your project ID\n",
+        "\n",
+        "**If you don't know your project ID**, you may be able to get your project ID using `gcloud`."
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,

notebooks/community/ml_ops/stage1/mlops_data_management.ipynb

@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 1 : data management."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:bq,chicago,lbn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Chicago Taxi](https://www.kaggle.com/chicago/chicago-taxi-trips-bq). The version of the dataset you will use in this tutorial is stored in a public BigQuery table. The trained model predicts whether someone would leave a tip for a taxi fare."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -120,6 +109,34 @@
         " - Preprocess the data with `Dataflow`"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:bq,chicago,lbn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Chicago Taxi](https://www.kaggle.com/chicago/chicago-taxi-trips-bq). The version of the dataset used in this tutorial is stored in a public BigQuery table. The trained model predicts whether someone leaves a tip for a taxi fare."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "9e483012a752"
+      },
+      "source": [
+        "### Costs\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "- Vertex AI\n",
+        "- Cloud Storage\n",
+        "- BigQuery\n",
+        "- Dataflow\n",
+        "\n",
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing), [BigQuery pricing](https://cloud.google.com/bigquery/pricing), and [Dataflow pricing](https://cloud.google.com/dataflow/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -142,7 +159,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -154,18 +171,19 @@
         "\n",
         "ONCE_ONLY = True\n",
         "if ONCE_ONLY:\n",
-        "    ! pip3 install -U tensorflow==2.5 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-data-validation==1.2 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-transform==1.2 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-io==0.18 $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-pipeline-components $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-bigquery $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-logging $USER_FLAG\n",
-        "    ! pip3 install --upgrade apache-beam[gcp] $USER_FLAG\n",
-        "    ! pip3 install --upgrade pyarrow $USER_FLAG\n",
-        "    ! pip3 install --upgrade cloudml-hypertune $USER_FLAG\n",
-        "    ! pip3 install --upgrade kfp $USER_FLAG"
+        "    ! pip3 install -U tensorflow==2.5 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-data-validation==1.2 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-transform==1.2 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-io==0.18 $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-pipeline-components $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-bigquery $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-logging $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade apache-beam[gcp]==2.33.0 $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade pyarrow $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade cloudml-hypertune $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade kfp $USER_FLAG -q\n",
+        "    ! pip3 install future $USER_FLAG -q"
       ]
     },
     {
@@ -373,12 +391,11 @@
         "import sys\n",
         "\n",
         "# If on Vertex AI Workbench, then don't execute this code\n",
-        "IS_COLAB = False\n",
+        "IS_COLAB = \"google.colab\" in sys.modules\n",
         "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
         "    \"DL_ANACONDA_HOME\"\n",
         "):\n",
         "    if \"google.colab\" in sys.modules:\n",
-        "        IS_COLAB = True\n",
         "        from google.colab import auth as google_auth\n",
         "\n",
         "        google_auth.authenticate_user()\n",
@@ -649,7 +666,7 @@
       },
       "outputs": [],
       "source": [
-        "bqclient = bigquery.Client()"
+        "bqclient = bigquery.Client(project=PROJECT_ID)"
       ]
     },
     {
@@ -772,6 +789,11 @@
         "LIMIT = 300000\n",
         "YEAR = 2020\n",
         "\n",
+        "# First, create the dataset entry\n",
+        "dataset = bigquery.Dataset(f\"{PROJECT_ID}.{BQ_DATASET}\")\n",
+        "dataset.location = \"US\"\n",
+        "dataset = bqclient.create_dataset(dataset, timeout=30)\n",
+        "\n",
         "query = f\"\"\"\n",
         "CREATE OR REPLACE TABLE `{BQ_TABLE_COPY}`\n",
         "AS (\n",
@@ -1212,7 +1234,7 @@
         "import setuptools\n",
         "\n",
         "REQUIRED_PACKAGES = [\n",
-        "    \"google-cloud-aiplatform==1.4.2\",\n",
+        "    \"google-cloud-aiplatform\",\n",
         "    \"tensorflow-transform==1.2.0\",\n",
         "    \"tensorflow-data-validation==1.2.0\",\n",
         "]\n",

notebooks/community/ml_ops/stage2/README.md

@@ -35,79 +35,144 @@ The second stage in MLOps is experimenting in developing one or more baseline mo
 
 ### Get Started
 
-[Get Started with Vertex Experiments and Vertex ML Metadata](get_started_vertex_experiments.ipynb)
+
+[Get started with Vertex AI Training for Pytorch](get_started_vertex_training_pytorch.ipynb)
 
 ```
 The steps performed include:
-
-- Use Python logging to log training configuration/results locally.
-- Use Google Cloud Logging to log training configuration/results in cloud storage.
-- Create a Vertex AI `Experiment` resource.
-- Instantiate an experiment run.
-- Log parameters for the run.
-- Log metrics for the run.
-- Display the logged experiment run.
-```
-
-[Get Started with Vertex TensorBoard](get_started_vertex_tensorboard.ipynb)
-
-```
-The steps performed include:
-
-- Create a TensorBoard callback when training a model.
-- Using Tensorboard with locally trained model.
-- Using Vertex AI TensorBoard with Vertex AI Training.
-```
-
-[Get Started with Custom Training Packages (Tensorflow)](get_started_vertex_training.ipynb)
-
-```
-The steps performed include:
-
-- Training using a single Python script.
-- Training using a Python package.
-- Training using a custom training image.
-- Laying out a training package.
-```
-
-[Get Started with Custom Training Packages (Scikit-Learn)](get_started_vertex_training_sklearn.ipynb)
-
-```
-The steps performed include:
-
-- Training using a Python package.
-- Report accuracy when hyperparameter tuning.
-- Save the model artifacts to Cloud Storage using GCSFuse.
-- Create a `Vertex AI Model` resource.
-```
-
-[Get Started with Custom Training Packages (XGBoost)](get_started_vertex_training_xgboost.ipynb)
-
-```
-The steps performed include:
-
-- Training using a Python package.
-- Report accuracy when hyperparameter tuning.
-- Save the model artifacts to Cloud Storage using GCSFuse.
-- Create a `Vertex AI Model` resource.
-```
-
-[Get Started with Custom Training Packages (Pytorch)](get_started_vertex_training_pytorch.ipynb)
-
-```
-The steps performed include:
-
 - Single node training using a Python package.
 - Report accuracy when hyperparameter tuning.
 - Save the model artifacts to Cloud Storage using GCSFuse.
 - Create a `Vertex AI Model` resource.
 ```
 
-[Get Started with Custom Training Packages (R)](get_started_vertex_training_r.ipynb)
+[Get started with prebuilt TFHub models](get_started_with_tfhub_models.ipynb)
 
 ```
 The steps performed include:
+- Download a TensorFlow Hub prebuilt model.
+- Add the task component as a classifier for the CIFAR-10 dataset.
+- Fine tune locally the model with transfer learning training.
+- Construct a custom training script:
+    - Get training data from TensorFlow Datasets
+    - Get model architecture from TensorFlow Hub
+    - Train then model
+    - Save model artifacts and upload as Vertex AI Model resource.
+```
 
+[Get started with Vertex AI TensorBoard](get_started_vertex_tensorboard.ipynb)
+
+```
+The steps performed include:
+- Create a TensorBoard callback when training a model.
+- Using Tensorboard with locally trained model.
+- Using Vertex AI TensorBoard with Vertex AI Training.
+```
+
+[Get started with TabNet builtin algorithm for training tabular models](get_started_with_tabnet.ipynb)
+
+```
+The steps performed include:
+- Get the training data.
+- Configure training parameters for the `Vertex AI TabNet` container.
+- Train the model using `Vertex AI Training` using CSV data.
+- Upload the model as a `Vertex AI Model` resource.
+- Deploy the `Vertex AI Model` resource to a `Vertex AI Endpoint` resource.
+- Make a prediction with the deployed model.
+- Hyperparameter tuning the `Vertex AI TabNet` model.
+- Train the model using `Vertex AI Training` using BigQuery table.
+
+```
+
+[Get started with Vertex AI Vizier](get_started_vertex_vizier.ipynb)
+
+```
+The steps performed include:
+- Hyperparameter tuning with Random algorithm.
+- Hyperparameter tuning with Vizier (Bayesian) algorithm.
+```
+
+[Automl image classfication training with customer managed encryption keys (CMEK)](get_started_with_cmek_training.ipynb)
+
+```
+The steps performed include:
+- Creating a customer managed encryption key.
+- Creating an image dataset with CMEK encryption.
+- Train an AutoML model with CMEK encryption.
+```
+
+[Get started with Vertex AI distributed training](get_started_vertex_distributed_training.ipynb)
+
+```
+The steps performed include:
+- `MirroredStrategy`: Train on a single VM with multiple GPUs.
+- `MultiWorkerMirroredStrategy`: Train on multiple VMs with automatic setup of replicas.
+- `MultiWorkerMirroredStrategy`: Train on multiple VMs with fine grain control of replicas.
+- `ReductionServer`: Train on multiple VMS and sync updates across VMS with `Vertex AI Reduction Server`.
+- `TPUTraining`: Train with multiple Cloud TPUs.
+```
+
+[Get started with Vertex AI Training for scikit-learn](get_started_vertex_training_sklearn.ipynb)
+
+```
+The steps performed include:
+- Training using a Python package.
+- Report accuracy when hyperparameter tuning.
+- Save the model artifacts to Cloud Storage using GCSFuse.
+- Create a `Vertex AI Model` resource.
+```
+
+[Get started with Vertex AI Experiments](get_started_vertex_experiments.ipynb)
+
+```
+The steps performed include:
+- Local (notebook) Training
+    - Create an experiment
+    - Create a first run in the experiment
+    - Log parameters and metrics
+    - Create artifact lineage
+    - Visualize the experiment results
+    - Execute a second run
+    - Compare the two runs in the experiment
+- Cloud (`Vertex AI`) Training
+    - Within the training script:
+        - Create an experiment
+        - Log parameters and metrics
+        - Create artifact lineage
+    - Create a `Vertex AI Training` custom job
+    - Execute the custom job
+    - Visualize the experiment results
+```
+
+[Get started with Vertex AI Hyperparameter Tuning for XGBoost](get_started_vertex_hpt_xgboost.ipynb)
+
+```
+The steps performed include:
+- Training using a Python package.
+- Report accuracy when hyperparameter tuning.
+- Save the model artifacts to Cloud Storage using GCSFuse.
+- Create a `Vertex AI Model` resource.
+
+```
+
+[Get started with Vertex AI Feature Store](get_started_vertex_feature_store.ipynb)
+
+```
+The steps performed include:
+- Creating a Vertex AI `Featurestore` resource.
+    - Creating `EntityType` resources for the `Featurestore` resource.
+    - Creating `Feature` resources for each `EntityType` resource.
+- Import feature values (entity data items) into `Featurestore` resource.
+    - From a Cloud Storage location.
+    - From a pandas DataFrame.
+- Perform online serving from a `Featurestore` resource.
+- Perform batch serving from a `Featurestore` resource.
+```
+
+[Get started with Vertex AI Training for R](get_started_vertex_training_r.ipynb)
+
+```
+The steps performed include:
 - Locally train an R model in a notebook using %%R magic commands
 - Create a deployment image with trained R model and serving functions.
 - Test the deployment image locally.
@@ -119,70 +184,32 @@ The steps performed include:
 - Train a R model using `Vertex AI Trainingh` service with the R-to-Python training package.
 ```
 
-[Get Started with Custom Training Packages (R) and Deployment in R environment](get_started_vertex_training_r_using_r_kernel.ipynb)
+[Get started with logging](get_started_with_logging.ipynb)
+
 ```
 The steps performed include:
+- Use Python logging to log training configuration/results locally.
+- Use Google Cloud Logging to log training configuration/results in cloud storage.
+```
 
+[Get started with Vertex AI Training for R using R Kernel](get_started_vertex_training_r_using_r_kernel.ipynb)
+
+```
+The steps performed include:
 - Create a custom R training script
 - Create a custom R serving script
 - Create a custom R deployment (serving) container.
 - Train the model using `Vertex AI` custom training.
-- Create an `Endpoint` resource.
+- Create an `Endpoint` resouce.
 - Deploy the `Model` resource (trained R model) to the `Endpoint` resource.
 - Make an online prediction.
+
 ```
 
-[Get Started with Custom Training Packages (LightGBM)](get_started_vertex_training_lightgbm.ipynb)
+[Get started with BigQuery ML training](get_started_bqml_training.ipynb)
 
 ```
 The steps performed include:
-
-- Training using a Python package.
-- Save the model artifacts to Cloud Storage using GCSFuse.
-- Construct a FastAPI prediction server.
-- Construct a Dockerfile deployment image.
-- Test the deployment image locally.
-- Create a `Vertex AI Model` resource.
-```
-
-[Get Started with Distributed Training](get_started_vertex_distributed_training.ipynb)
-
-```
-The steps performed include:
-
-- `MirroredStrategy`: Train on a single VM with multiple GPUs.
-- `MultiWorkerMirroredStrategy`: Train on multiple VMs with automatic setup of replicas.
-- `MultiWorkerMirroredStrategy`: Train on multiple VMs with fine grain control of replicas.
-- `ReductionServer`: Train on multiple VMS and sync updates across VMS with `Vertex AI Reduction Server`.
-- `TPUTraining`: Train with multiple Cloud TPUs.
-```
-
-[Get Started with Vizier Hyperparameter Tuning](get_started_vertex_vizier.ipynb)
-
-```
-The steps performed include:
-
-- Hyperparameter tuning with Random algorithm.
-- Hyperparameter tuning with Vizier (Bayesian) algorithm.
-```
-
-[Get Started with AutoML Training](get_started_automl_training.ipynb)
-
-```
-The steps performed include:
-
-- Train an image model.
-- Export the image model as an edge model.
-- Train a tabular model.
-- Export the tabular model as a cloud model.
-- Train a text model.
-```
-
-[Get Started with BQML Training](get_started_bqml_training.ipynb)
-
-```
-The steps performed include:
-
 - Create a local BigQuery table in your project
 - Train a BQML model
 - Evaluate the BQML model
@@ -190,65 +217,59 @@ The steps performed include:
 - Upload the exported model as a `Vertex AI Model` resource
 - Hyperparameter tune a BQML model with `Vertex AI Vizier`
 - Automatically register a BQML model to `Vertex AI Model Registry`
+
 ```
 
-[Get Started with Vertex Feature Store](get_started_vertex_feature_store.ipynb)
+[Get started with AutoML training](get_started_automl_training.ipynb)
 
 ```
 The steps performed include:
-
-- Creating a Vertex AI `Featurestore` resource.
-    - Creating `EntityType` resources for the `Featurestore` resource.
-    - Creating `Feature` resources for each `EntityType` resource.
-- Import feature values (entity data items) into `Featurestore` resource from Cloud Storage.
-- Import feature values (entity data items) into `Featurestore` resource from pandas DataFrame.
-- Perform online serving from a `Featurestore` resource.
-- Perform batch serving from a `Featurestore` resource.
+- Train an image model
+- Export the image model as an edge model
+- Train a tabular model
+- Export the tabular model as a cloud model
+- Train a text model
+- Train a video model
 ```
 
-[Get Started with Google CMEK Training](get_started_with_cmek_training.ipynb)
+[Get started with Vertex AI Training for XGBoost](get_started_vertex_training_xgboost.ipynb)
 
 ```
 The steps performed include:
+- Training using a Python package.
+- Report accuracy when hyperparameter tuning.
+- Save the model artifacts to Cloud Storage using GCSFuse.
+- Create a `Vertex AI Model` resource.
 
-- Creating a customer managed encryption key.
-- Creating an image dataset with CMEK encryption.
-- Train an AutoML model with CMEK encryption.
 ```
 
-[Get Started with TensorFlow Hub models](get_started_with_tfhub_models.ipynb)
+[Get started with Vertex AI Training](get_started_vertex_training.ipynb)
 
 ```
 The steps performed include:
-
-- Download a TensorFlow Hub prebuilt model.
-- Add the task component as a classifier for the CIFAR-10 dataset.
-- Fine tune locally the model with transfer learning training.
-- Construct a custom training script:
-    - Get training data from TensorFlow Datasets
-    - Get model architecture from TensorFlow Hub
-    - Train then model
-    - Save model artifacts and upload as Vertex AI Model resource.
+- Training using a single Python script.
+- Training using a Python package.
+- Training using a custom training image.
+- Laying out a training package.
 ```
 
-[Get Started with Vertex AI TabNet builtin algorithm](get_started_with_tabnet.ipynb)
+[Get started with Vertex AI Training for LightGBM](get_started_vertex_training_lightgbm.ipynb)
+
 ```
 The steps performed include:
+- Training using a Python package.
+- Save the model artifacts to Cloud Storage using GCSFuse.
+- Construct a FastAPI prediction server.
+- Construct a Dockerfile deployment image.
+- Test the deployment image locally.
+- Create a `Vertex AI Model` resource.
 
-- Get the training data.
-- Configure training parameters for the Vertex AI TabNet container.
-- Train the model using Vertex AI Training using CSV data.
-- Upload the model as a Vertex AI Model resource.
-- Deploy the Vertex AI Model resource to a Vertex AI Endpoint resource.
-- Make a prediction with the deployed model.
-- Hyperparameter tuning the Vertex AI TabNet model.
-- Train the model using Vertex AI Training using BigQuery table.
 ```
 
-[Get Started with Vision API and AutoML](get_started_with_visionapi_and_automl.ipynb)
+[Get started Vision API test preprocessing and AutoML text model generation](get_started_with_visionapi_and_automl.ipynb)
+
 ```
 The steps performed include:
-
 - Preprocess training files using `Vision AI` APIs to extract the text from PDF files.
 - Create a custom import file that includes annotation data based on the sample `BigQuery` dataset.
 - Create a `Vertex AI Dataset` resource.
@@ -259,14 +280,12 @@ The steps performed include:
 - Undeploy the `Model`.
 ```
 
-
 ### E2E Stage Example
 
 [Stage 2: Experimentation](mlops_experimentation.ipynb)
 
 ```
 The steps performed include:
-
 - Review the `Dataset` resource created during stage 1.
 - Train an AutoML tabular binary classifier model in the background.
 - Build the experimental model architecture.

notebooks/community/ml_ops/stage2/get_started_automl_training.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_automl_training.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -65,52 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with AutoML Training."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Datasets\n",
-        "\n",
-        "#### Image\n",
-        "\n",
-        "The image dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower in a given image from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:gsod,lrg"
-      },
-      "source": [
-        "#### Tabular\n",
-        "\n",
-        "The tabular dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:happydb,tcn"
-      },
-      "source": [
-        "#### Text\n",
-        "\n",
-        "The text dataset used for this tutorial is the [Happy Moments dataset](https://www.kaggle.com/ritresearch/happydb) from [Kaggle Datasets](https://www.kaggle.com/ritresearch/happydb). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "98eb93ec6faa"
-      },
-      "source": [
-        "#### Video\n",
-        "\n",
-        "The video dataset used for this tutorial is the golf swing recognition portion of the [Human Motion dataset](https://todo) from [MIT](http://cbcl.mit.edu/publications/ps/Kuehne_etal_iccv11.pdf). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model will predict the start frame where a golf swing begins."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -150,6 +104,31 @@
         "* **You want to establish a baseline metric before experimenting with a custom model**"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Datasets\n",
+        "\n",
+        "#### Image\n",
+        "\n",
+        "The image dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower in a given image from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip.\n",
+        "\n",
+        "#### Tabular\n",
+        "\n",
+        "The tabular dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp).\n",
+        "\n",
+        "#### Text\n",
+        "\n",
+        "The text dataset used for this tutorial is the [Happy Moments dataset](https://www.kaggle.com/ritresearch/happydb) from [Kaggle Datasets](https://www.kaggle.com/ritresearch/happydb). The version of the dataset you use in this tutorial is stored in a public Cloud Storage bucket.\n",
+        "\n",
+        "#### Video\n",
+        "\n",
+        "The video dataset used for this tutorial is the golf swing recognition portion of the [Human Motion dataset](https://todo) from [MIT](http://cbcl.mit.edu/publications/ps/Kuehne_etal_iccv11.pdf). The version of the dataset you use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the start frame where a golf swing begins."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -187,7 +166,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -238,6 +217,8 @@
         "id": "project_id"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -253,8 +234,15 @@
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
         "\n",
-        "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands.\n",
-        "\n",
+        "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "5aee4379e8e5"
+      },
+      "source": [
         "#### Set your project ID\n",
         "\n",
         "**If you don't know your project ID**, you may be able to get your project ID using `gcloud`."

notebooks/community/ml_ops/stage2/get_started_bqml_training.ipynb

@@ -40,11 +40,11 @@
         "  </td>\n",
         "    <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_bqml_training.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_bqml_training.ipynb\">\n",
+        "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/notebooks/community/ml_ops/stage2/get_started_bqml_training.ipynb\">\n",
         "      <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
         "      Open in Vertex AI Workbench\n",
         "    </a>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with BigQuery ML Training."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:penguins,lcn,bq"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the Penguins dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). This version of the dataset is used to predict the species of penguins from the available features like culmen-length, flipper-depth etc."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -100,8 +89,26 @@
         "- Export the BQML model as a cloud model\n",
         "- Upload the exported model as a `Vertex AI Model` resource\n",
         "- Hyperparameter tune a BQML model with `Vertex AI Vizier`\n",
-        "- Automatically register a BQML model to `Vertex AI Model Registry`\n",
+        "- Automatically register a BQML model to `Vertex AI Model Registry`"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:penguins,lcn,bq"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is the Penguins dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). This version of the dataset is used to predict the species of penguins from the available features like culmen-length, flipper-depth etc."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "81c777b8ad32"
+      },
+      "source": [
         "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -134,7 +141,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\"/opt/deeplearning/metadata/env_version\")\n",
         "\n",
         "# Vertex AI Notebook requires dependencies to be installed with '--user'\n",
@@ -184,6 +191,8 @@
         "id": "project_id"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -199,8 +208,15 @@
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
         "\n",
-        "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands.\n",
-        "\n",
+        "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "56d591439df1"
+      },
+      "source": [
         "#### Set your project ID\n",
         "\n",
         "**If you don't know your project ID**, you may be able to get your project ID using `gcloud`."
@@ -442,6 +458,54 @@
         "! gsutil ls -al $BUCKET_URI"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "set_service_account"
+      },
+      "source": [
+        "#### Service Account\n",
+        "\n",
+        "You use a service account to create Vertex AI Pipeline jobs. If you do not want to use your project's Compute Engine service account, set `SERVICE_ACCOUNT` to another service account ID."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "set_service_account"
+      },
+      "outputs": [],
+      "source": [
+        "SERVICE_ACCOUNT = \"[your-service-account]\"  # @param {type:\"string\"}"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "autoset_service_account"
+      },
+      "outputs": [],
+      "source": [
+        "if (\n",
+        "    SERVICE_ACCOUNT == \"\"\n",
+        "    or SERVICE_ACCOUNT is None\n",
+        "    or SERVICE_ACCOUNT == \"[your-service-account]\"\n",
+        "):\n",
+        "    # Get your service account from gcloud\n",
+        "    if not IS_COLAB:\n",
+        "        shell_output = !gcloud auth list 2>/dev/null\n",
+        "        SERVICE_ACCOUNT = shell_output[2].replace(\"*\", \"\").strip()\n",
+        "\n",
+        "    if IS_COLAB:\n",
+        "        shell_output = ! gcloud projects describe  $PROJECT_ID\n",
+        "        project_number = shell_output[-1].split(\":\")[1].strip().replace(\"'\", \"\")\n",
+        "        SERVICE_ACCOUNT = f\"{project_number}-compute@developer.gserviceaccount.com\"\n",
+        "\n",
+        "    print(\"Service Account:\", SERVICE_ACCOUNT)"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -730,32 +794,6 @@
         "print(\"{} created in {}\".format(tblname, job.ended - job.started))"
       ]
     },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "3b3aa4481cd7"
-      },
-      "outputs": [],
-      "source": [
-        "MODEL_QUERY = f\"\"\"\n",
-        "DROP MODEL `{BQ_DATASET_NAME}.{MODEL_NAME}`\n",
-        "\"\"\"\n",
-        "\n",
-        "job = bqclient.query(MODEL_QUERY)"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "ef9e14b91475"
-      },
-      "outputs": [],
-      "source": [
-        "job"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -1185,24 +1223,21 @@
         "\n",
         "### Setting permissions to automatically register the model\n",
         "\n",
-        "You need to set some additional IAM permissions for BigQuery ML to automatically upload and register the model after training. Depending on your service account, the setting of the permissions below may fail. In this case, we recommend executing the permissions in a Cloud Shell."
+        "You need to set some additional IAM permissions for BigQuery ML to automatically upload and register the model after training. Depending on your service account, the setting of the permissions below may fail. In this case, we recommend executing the permissions in a Cloud Shell.\n",
+        "\n",
+        "Learn more about [Setting permissions for Model Registry](https://cloud.google.com/bigquery-ml/docs/managing-models-vertex)\n"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "0472e888105e"
+        "id": "29229f72d13d"
       },
       "outputs": [],
       "source": [
         "! gcloud projects add-iam-policy-binding $PROJECT_ID \\\n",
-        "      --member='serviceAccount:cloud-dataengine@system.gserviceaccount.com' \\\n",
-        "      --role='roles/aiplatform.admin'\n",
-        "\n",
-        "! gcloud projects add-iam-policy-binding $PROJECT_ID \\\n",
-        "      --member='user:cloud-dataengine@prod.google.com' \\\n",
-        "      --role='roles/aiplatform.admin'"
+        "    --member=serviceAccount:$SERVICE_ACCOUNT --role=roles/aiplatform.admin --condition=None"
       ]
     },
     {
@@ -1283,6 +1318,20 @@
         "print(model.gca_resource)"
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "48e6ef5d5ffa"
+      },
+      "outputs": [],
+      "source": [
+        "models = aiplatform.Model.list()\n",
+        "for model in models:\n",
+        "    if model.gca_resource.display_name.startswith(\"bqml\"):\n",
+        "        print(model.gca_resource.display_name)"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {

notebooks/community/ml_ops/stage2/get_started_vertex_distributed_training.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex Distributed Training\n",
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI Distributed Training\n",
         "\n",
         "<table align=\"left\">\n",
         "      <td>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex Distributed Training. Please note: There are incompatibilities between Colab and Docker and the Docker section may not work until resolved by the platform."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:custom,boston,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Boston Housing Prices dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html). The version of the dataset you will use in this tutorial is built into TensorFlow. The trained model predicts the median price of a house in units of 1K USD."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -106,15 +95,6 @@
         "id": "recommendation:mlops,stage2,vertex,distributed_training"
       },
       "source": [
-        "### Costs\n",
-        " \n",
-        "This tutorial uses billable components of Google Cloud:\n",
-        "\n",
-        "Vertex AI\n",
-        "Cloud Storage\n",
-        "\n",
-        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage pricing](https://cloud.google.com/storage/pricing), and use the [Pricing Calculator](https://cloud.google.com/products/calculator/),\n",
-        "        to generate a cost estimate based on your projected usage.\n",
         "### Recommendations\n",
         "\n",
         "When doing E2E MLOps on Google Cloud, the following are best practices for when to use Vertex AI Distributed Training:\n",
@@ -138,13 +118,41 @@
         "While training across a large number of VMs and the model parameters updates to sync is very large."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:custom,boston,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Boston Housing Prices dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html). The version of the dataset you use in this tutorial is built into TensorFlow. The trained model predicts the median price of a house in units of 1K USD."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "d10166df7141"
+      },
+      "source": [
+        "### Costs\n",
+        " \n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "Vertex AI\n",
+        "Cloud Storage\n",
+        "\n",
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage pricing](https://cloud.google.com/storage/pricing), and use the [Pricing Calculator](https://cloud.google.com/products/calculator/),\n",
+        "        to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "XkYpRvOQyVYb"
       },
       "source": [
-        "### Install additional packages\n",
+        "## Installation\n",
         "\n",
         "Install the packages required for executing this notebook."
       ]
@@ -160,7 +168,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -170,7 +178,7 @@
         "if IS_WORKBENCH_NOTEBOOK:\n",
         "    USER_FLAG = \"--user\"\n",
         "\n",
-        "! pip3 install {USER_FLAG} --upgrade google-cloud-aiplatform"
+        "! pip3 install {USER_FLAG} --upgrade google-cloud-aiplatform -q"
       ]
     },
     {
@@ -248,8 +256,6 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
         "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },

notebooks/community/ml_ops/stage2/get_started_vertex_feature_store.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Feature Store\n",
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI Feature Store\n",
         "\n",
         "<table align=\"left\">\n",
         "  <td>\n",
@@ -41,7 +41,7 @@
         "    \n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_feature_store.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "    \n",
@@ -68,19 +68,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Feature Store."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:movies,lbn,avro"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the `Movie Recommendations` dataset. The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket, in Avro format.\n",
-        "\n",
-        "This dataset is used to predict whether a person will watch a movie or not."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -107,6 +94,19 @@
         "- Perform batch serving from a `Featurestore` resource."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:movies,lbn,avro"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the `Movie Recommendations` dataset. The version of the dataset you use in this tutorial is stored in a public Cloud Storage bucket, in Avro format.\n",
+        "\n",
+        "This dataset is used to predict whether a person watches a movie or not."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -145,7 +145,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -194,6 +194,8 @@
         "id": "project_id"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -362,12 +364,11 @@
         "import sys\n",
         "\n",
         "# If on Vertex AI Workbench, then don't execute this code\n",
-        "IS_COLAB = False\n",
+        "IS_COLAB = \"google.colab\" in sys.modules\n",
         "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
         "    \"DL_ANACONDA_HOME\"\n",
         "):\n",
         "    if \"google.colab\" in sys.modules:\n",
-        "        IS_COLAB = True\n",
         "        from google.colab import auth as google_auth\n",
         "\n",
         "        google_auth.authenticate_user()\n",
@@ -490,7 +491,7 @@
       "outputs": [],
       "source": [
         "# Represents featurestore resource path.\n",
-        "FEATURESTORE_NAME = \"movies\"\n",
+        "FEATURESTORE_NAME = \"movies_\" + TIMESTAMP\n",
         "\n",
         "featurestore = aiplatform.Featurestore.create(\n",
         "    featurestore_id=FEATURESTORE_NAME,\n",

notebooks/community/ml_ops/stage2/get_started_vertex_tensorboard.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI Tensorboard\n",
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI TensorBoard\n",
         "\n",
         "<table align=\"left\">\n",
         "   <td>\n",
@@ -62,7 +62,7 @@
         "## Overview\n",
         "\n",
         "\n",
-        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Tensorboard."
+        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI TensorBoard."
       ]
     },
     {
@@ -83,10 +83,44 @@
         "The steps performed include:\n",
         "\n",
         "- Create a TensorBoard callback when training a model.\n",
-        "- Using Tensorboard with locally trained model.\n",
+        "- Using TensorBoard with locally trained model.\n",
         "- Using Vertex AI TensorBoard with Vertex AI Training."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "recommendation:mlops,stage2,vertex,tensorboard"
+      },
+      "source": [
+        "### Recommendations\n",
+        "\n",
+        "When doing E2E MLOps on Google Cloud, the following are the best practices for visualizing your training with TensorBoard.\n",
+        "\n",
+        "#### Local TensorBoard\n",
+        "\n",
+        "Use the OSS version of TensorBoard, either command-line or daemon version, when doing ad-hoc training locally.\n",
+        "\n",
+        "#### Cloud TensorBoard\n",
+        "\n",
+        "Use the tensorboard.dev, when doing training on the cloud -- unless you have a privacy issue.\n",
+        "\n",
+        "#### Experiments\n",
+        "\n",
+        "Use Vertex AI TensorBoard when you have a privacy issue or doing experiments to compare results for different experiment configurations."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "03bfd1274241"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "In this tutorial you use the MNIST dataset. The version of the dataset is built into the TF.Keras framework. The dataset predicts which digit an image is, between 0 .. 9."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -115,8 +149,8 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "**If you are using Colab or Google Cloud Notebooks**, your environment already meets\n",
-        "all the requirements to run this notebook. You can skip this step.\n",
+        "**If you are using Colab or Vertex AI Workbench Notebooks**, your environment already meets\n",
+        "all the requirements to run this notebook.  \n",
         "\n",
         "**Otherwise**, make sure your environment meets this notebook's requirements.\n",
         "You need the following:\n",
@@ -149,29 +183,6 @@
         "1. Open this notebook in the Jupyter Notebook Dashboard.\n"
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "recommendation:mlops,stage2,vertex,tensorboard"
-      },
-      "source": [
-        "### Recommendations\n",
-        "\n",
-        "When doing E2E MLOps on Google Cloud, the following are the best practices for visualizing your training with TensorBoard.\n",
-        "\n",
-        "#### Local TensorBoard\n",
-        "\n",
-        "Use the OSS version of TensorBoard, either command-line or daemon version, when doing ad-hoc training locally.\n",
-        "\n",
-        "#### Cloud TensorBoard\n",
-        "\n",
-        "Use the Tensorboard.dev, when doing training on the cloud -- unless you have a privacy issue.\n",
-        "\n",
-        "#### Experiments\n",
-        "\n",
-        "Use Vertex AI TensorBoard when you have a privacy issue or doing experiments to compare results for different experiment configurations."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -194,7 +205,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -255,7 +266,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -774,9 +785,9 @@
       "source": [
         "## Training with TensorBoard\n",
         "\n",
-        "Tensorboard provides the means to visualize your training in-real time and to visualize the results (metrics).\n",
+        "TensorBoard provides the means to visualize your training in-real time and to visualize the results (metrics).\n",
         "\n",
-        "You can use Tensorboard in conjunction with local training, cloud training and with `Vertex AI Training`, which is referred to as `Vertex AI TensorBoard`"
+        "You can use TensorBoard in conjunction with local training, cloud training and with `Vertex AI Training`, which is referred to as `Vertex AI TensorBoard`"
       ]
     },
     {

notebooks/community/ml_ops/stage2/get_started_vertex_training.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -62,18 +62,7 @@
         "## Overview\n",
         "\n",
         "\n",
-        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex Training."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:custom,boston,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Boston Housing Prices dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html). The version of the dataset you will use in this tutorial is built into TensorFlow. The trained model predicts the median price of a house in units of 1K USD."
+        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Training."
       ]
     },
     {
@@ -125,6 +114,38 @@
         "CustomJob"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:custom,boston,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Boston Housing Prices dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html). The version of the dataset this tutorial is built into TensorFlow. The trained model predicts the median price of a house in units of 1K USD."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "c480fc50ec3c"
+      },
+      "source": [
+        "### Costs \n",
+        "\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "* Vertex AI\n",
+        "* Cloud Storage\n",
+        "\n",
+        "\n",
+        "Learn about [Vertex AI\n",
+        "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
+        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
+        "Calculator](https://cloud.google.com/products/calculator/)\n",
+        "to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -147,7 +168,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -209,7 +230,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_vertex_training_lightgbm.ipynb

@@ -29,7 +29,7 @@
         "id": "4f82ca678df6"
       },
       "source": [
-        "Notebook is a revised version of notebook from [Rajesh Thallam](https://github.com/RajeshThallam/vertex-ai-labs/blob/main/07-vertex-train-deploy-lightgbm/vertex-train-deploy-lightgbm-model.ipynb)"
+        "This notebook is a revised version of notebook from [Rajesh Thallam](https://github.com/RajeshThallam/vertex-ai-labs/blob/main/07-vertex-train-deploy-lightgbm/vertex-train-deploy-lightgbm-model.ipynb)"
       ]
     },
     {
@@ -43,12 +43,12 @@
         "<table align=\"left\">\n",
         "\n",
         "  <td>\n",
-        "    <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/master/notebooks/community/ml_ops/stage2/get_started_vertex_training_lightgbm.ipynb\">\n",
+        "    <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_lightgbm.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/master/notebooks/ocommunity/ml_ops/stage2/get_started_vertex_training_lightgbm.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_lightgbm.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
@@ -56,7 +56,7 @@
         "  <td>\n",
         "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_lightgbm.ipynb\">\n",
         "      <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
-        "      Run in Vertex Workbench\n",
+        "      Open in Vertex AI Workbench\n",
         "    </a>\n",
         "  </td>\n",
         "</table>"
@@ -74,17 +74,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Training for LightGBM."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:iris,lcn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Iris dataset](https://www.tensorflow.org/datasets/catalog/iris) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). This dataset does not require any feature engineering. The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of Iris flower species from a class of three species: setosa, virginica, or versicolor."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -107,8 +96,26 @@
         "- Construct a FastAPI prediction server.\n",
         "- Construct a Dockerfile deployment image.\n",
         "- Test the deployment image locally.\n",
-        "- Create a `Vertex AI Model` resource.\n",
+        "- Create a `Vertex AI Model` resource."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:iris,lcn"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is the [Iris dataset](https://www.tensorflow.org/datasets/catalog/iris) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). This dataset does not require any feature engineering. The version of the dataset in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of Iris flower species from a class of three species: setosa, virginica, or versicolor."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "de76bb18c85b"
+      },
+      "source": [
         "### Costs \n",
         "\n",
         "This tutorial uses billable components of Google Cloud:\n",
@@ -131,7 +138,7 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "If you are using Colab or Google Cloud Notebooks, your environment already meets all the requirements to run this notebook. You can skip this step.\n",
+        "If you are using Colab or Vertex AI Workbench Notebooks, your environment already meets all the requirements to run this notebook. You can skip this step.\n",
         "\n",
         "Otherwise, make sure your environment meets this notebook's requirements. You need the following:\n",
         "\n",
@@ -178,7 +185,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -190,19 +197,10 @@
         "\n",
         "! pip3 install --upgrade google-cloud-aiplatform $USER_FLAG -q\n",
         "! pip3 install -U google-cloud-storage $USER_FLAG -q\n",
-        "! pip3 install -U lightgbm $USER_FLAG -q"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "install_tensorflow"
-      },
-      "outputs": [],
-      "source": [
+        "! pip3 install -U lightgbm $USER_FLAG -q\n",
+        "\n",
         "if os.getenv(\"IS_TESTING\"):\n",
-        "    ! pip3 install --upgrade tensorflow $USER_FLAG"
+        "    ! pip3 install --upgrade tensorflow $USER_FLAG -q"
       ]
     },
     {
@@ -256,7 +254,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_vertex_training_pytorch.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex Training for Pytorch\n",
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI Training for Pytorch\n",
         "\n",
         "<table align=\"left\">\n",
         "  <td>\n",
@@ -62,18 +62,7 @@
         "## Overview\n",
         "\n",
         "\n",
-        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex Training for Pytorch."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:pytorch,cifar10,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [CIFAR10 dataset](https://pytorch.org/vision/stable/datasets.html#cifar) from [Pytorch Datasets](https://pytorch.org/vision/stable/datasets.html). The version of the dataset you will use is built into TensorFlow. The trained model predicts which type of class an image is from ten classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, or truck."
+        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Training for Pytorch."
       ]
     },
     {
@@ -100,13 +89,24 @@
         "- Create a `Vertex AI Model` resource."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:pytorch,cifar10,icn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [CIFAR10 dataset](https://pytorch.org/vision/stable/datasets.html#cifar) from [Pytorch Datasets](https://pytorch.org/vision/stable/datasets.html). The version of the dataset is built into TensorFlow. The trained model predicts which type of class an image is from ten classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, or truck."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "85ee859437ed"
       },
       "source": [
-        "## Costs \n",
+        "### Costs \n",
         "\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -128,7 +128,7 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "**If you are using Colab or Google Cloud Notebooks**, your environment already meets\n",
+        "**If you are using Colab or Vertex AI Workbench Notebooks**, your environment already meets\n",
         "all the requirements to run this notebook. You can skip this step."
       ]
     },
@@ -191,7 +191,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",

notebooks/community/ml_ops/stage2/get_started_vertex_training_r.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_r.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Training for R. Please note that this notebook should be ran only in R notebook image (e.g., R4.1)."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:r,iris,lcn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the Iris dataset built into the R package. This dataset does not require any feature engineering. The trained model predicts the type of Iris flower species from a class of three species: setosa, virginica, or versicolor."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -104,6 +93,17 @@
         "- Train a R model using `Vertex AI Trainingh` service with the R-to-Python training package."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:r,iris,lcn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the Iris dataset built into the R package. This dataset does not require any feature engineering. The trained model predicts the type of Iris flower species from a class of three species: setosa, virginica, or versicolor."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -148,7 +148,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -216,7 +216,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -243,8 +243,6 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
         "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },

notebooks/community/ml_ops/stage2/get_started_vertex_training_r_using_r_kernel.ipynb

@@ -29,7 +29,7 @@
         "id": "e3ba05e16cf2"
       },
       "source": [
-        "This is an updated version of a notebook contributed by [Fabian Hirschmann](https://github.com/fhirschmann)."
+        "This notebook is an updated version of a notebook contributed by [Fabian Hirschmann](https://github.com/fhirschmann)."
       ]
     },
     {
@@ -38,15 +38,17 @@
         "id": "JAPoU8Sm5E6e"
       },
       "source": [
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI Training for R using R Kernel\n",
+        "\n",
         "<table align=\"left\">\n",
         "\n",
         "  <td>\n",
-        "    <a href=\"https://colab.sandbox.google.com/github/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_r_using_r_kernel.ipynb\">\n",
+        "    <a href=\"https://colab.sandbox.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_r_using_r_kernel.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_r_using_r_kernel.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_training_r_using_r_kernel.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
@@ -63,18 +65,20 @@
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "tvgnzT1CKxrO"
+        "id": "be1799d4f500"
       },
       "source": [
         "## Overview\n",
         "\n",
-        "This example demonstrates how to train and deploy R models with `Vertex AI` using an R kernel -- such as in `Vertex AI Workbench Notebooks`.\n",
-        "\n",
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is [California Housing Dataset](https://www.dcc.fc.up.pt/~ltorgo/Regression/cal_housing.html). The data contains information from the 1990 California census. The data set is publicly available from Google Cloud Storage at `gs://cloud-samples-data/ai-platform-unified/datasets/tabular/california-housing-tabular-regression.csv`. The dataset is used to train a Random Forest regressor to predict a median housing price, given a longitude and lattitude along with data from the corresponding census block group. A block group is the smallest geographical unit for which the U.S. Census Bureau publishes sample data (a block group typically has a population of 600 to 3,000 people).\n",
-        "\n",
-        "\n",
+        "This example demonstrates how to train and deploy R models with `Vertex AI` using an R kernel -- such as in `Vertex AI Workbench Notebooks`."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "tvgnzT1CKxrO"
+      },
+      "source": [
         "### Objective\n",
         "\n",
         "In this tutorial, you learn how to use `Vertex AI`, using an R kernel, for training and deploying an R custom model.\n",
@@ -94,9 +98,26 @@
         "- Train the model using `Vertex AI` custom training.\n",
         "- Create an `Endpoint` resouce.\n",
         "- Deploy the `Model` resource (trained R model) to the `Endpoint` resource.\n",
-        "- Make an online prediction.\n",
-        "\n",
+        "- Make an online prediction.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "e1266da324d2"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is [California Housing Dataset](https://www.dcc.fc.up.pt/~ltorgo/Regression/cal_housing.html). The data contains information from the 1990 California census. The data set is publicly available from Google Cloud Storage at `gs://cloud-samples-data/ai-platform-unified/datasets/tabular/california-housing-tabular-regression.csv`. The dataset is used to train a Random Forest regressor to predict a median housing price, given a longitude and lattitude along with data from the corresponding census block group. A block group is the smallest geographical unit for which the U.S. Census Bureau publishes sample data (a block group typically has a population of 600 to 3,000 people).\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "de76bb18c85b"
+      },
+      "source": [
         "### Costs \n",
         "\n",
         "This tutorial uses billable components of Google Cloud:\n",
@@ -216,10 +237,10 @@
       },
       "outputs": [],
       "source": [
-        "required_packages <- c(\"reticulate\", \"glue\", \"httr\")\n",
+        "required_packages < -c(\"reticulate\", \"glue\", \"httr\")\n",
         "install.packages(setdiff(required_packages, rownames(installed.packages())))\n",
         "\n",
-        "sh(\"pip install --upgrade google-cloud-aiplatform\")"
+        "sh(\"pip3 install --upgrade google-cloud-aiplatform -q\")"
       ]
     },
     {
@@ -247,7 +268,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com) and the [Artifact Registry API](https://console.cloud.google.com/flows/enableapi?apiid=artifactregistry.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Have the project ID autodetected or enter it below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook."
@@ -272,7 +293,7 @@
       },
       "outputs": [],
       "source": [
-        "PROJECT_ID <- \"[your-project-id]\"  # @param {type:\"string\"}"
+        "PROJECT_ID < -\"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },
     {
@@ -440,8 +461,8 @@
       },
       "outputs": [],
       "source": [
-        "BUCKET_NAME <- \"[your-bucket-name]\"  # @param {type:\"string\"}\n",
-        "BUCKET_URI <- paste0(\"gs://\", BUCKET_NAME)"
+        "BUCKET_NAME < -\"[your-bucket-name]\"  # @param {type:\"string\"}\n",
+        "BUCKET_URI < -paste0(\"gs://\", BUCKET_NAME)"
       ]
     },
     {
@@ -611,9 +632,11 @@
       },
       "outputs": [],
       "source": [
-        "PRIVATE_REPO <- \"my-docker-repo\"\n",
+        "PRIVATE_REPO < -\"my-docker-repo\"\n",
         "\n",
-        "sh(\"gcloud artifacts repositories create {PRIVATE_REPO} --repository-format=docker --location={REGION} --description=\\\"Docker repository\\\"\")\n",
+        "sh(\n",
+        "    'gcloud artifacts repositories create {PRIVATE_REPO} --repository-format=docker --location={REGION} --description=\"Docker repository\"'\n",
+        ")\n",
         "\n",
         "sh(\"gcloud artifacts repositories list\")"
       ]
@@ -659,11 +682,13 @@
       },
       "outputs": [],
       "source": [
-        "IMAGE_NAME <- \"vertex-r\"  # @param {type:\"string\"}\n",
-        "IMAGE_TAG <- \"latest\"  # @param {type:\"string\"}\n",
-        "IMAGE_URI <- glue(\"{REGION}-docker.pkg.dev/{PROJECT_ID}/{PRIVATE_REPO}/{IMAGE_NAME}:{IMAGE_TAG}\")\n",
+        "IMAGE_NAME < -\"vertex-r\"  # @param {type:\"string\"}\n",
+        "IMAGE_TAG < -\"latest\"  # @param {type:\"string\"}\n",
+        "IMAGE_URI < -glue(\n",
+        "    \"{REGION}-docker.pkg.dev/{PROJECT_ID}/{PRIVATE_REPO}/{IMAGE_NAME}:{IMAGE_TAG}\"\n",
+        ")\n",
         "\n",
-        "dir.create(\"src\", showWarnings = FALSE)"
+        "dir.create(\"src\", showWarnings=FALSE)"
       ]
     },
     {
@@ -1022,14 +1047,6 @@
       },
       "outputs": [],
       "source": [
-        "INSTANCES<-'{\"instances\": [{\"longitude\": -122.22, \"latitude\": 37.86, \"housing_median_age\": 21, \"total_rooms\": 7099, \"total_bedrooms\": 1106, \"population\": 2401, \"households\": 1138, \"median_income\": 8.3014}, \n",
-        "                         {\"longitude\": -122.24, \"latitude\": 37.85, \"housing_median_age\": 52, \"total_rooms\": 1467, \"total_bedrooms\": 190, \"population\": 496, \"households\": 177, \"median_income\": 7.2574}, \n",
-        "                         {\"longitude\": -122.25, \"latitude\": 37.85, \"housing_median_age\": 52, \"total_rooms\": 1274, \"total_bedrooms\": 235, \"population\": 558, \"households\": 219, \"median_income\": 5.6431}\n",
-        "                        ]}'\n",
-        "\n",
-        "library(jsonlite)\n",
-        "json_instances = toJSON(INSTANCES)\n",
-        "\n",
         "df <- read.csv(text=sh(\"gsutil cat {data_uri}\", intern = TRUE))\n",
         "head(df, 5)\n",
         "\n",
@@ -1059,20 +1076,22 @@
       },
       "outputs": [],
       "source": [
-        "url <- glue(\"https://{REGION}-aiplatform.googleapis.com/v1/{endpoint$resource_name}:predict\")\n",
-        "access_token <- sh(\"gcloud auth print-access-token\", intern = TRUE)\n",
+        "url < -glue(\n",
+        "    \"https://{REGION}-aiplatform.googleapis.com/v1/{endpoint$resource_name}:predict\"\n",
+        ")\n",
+        "access_token < -sh(\"gcloud auth print-access-token\", intern=TRUE)\n",
         "\n",
         "sh(\n",
         "    \"curl\",\n",
-        "    c(\"--tr-encoding\",\n",
-        "      \"-s\",\n",
-        "      \"-X POST\",\n",
-        "      glue(\"-H 'Authorization: Bearer {access_token}'\"),\n",
-        "      \"-H 'Content-Type: application/jsoin'\",\n",
-        "      url,\n",
-        "      glue(\"-d {json_instances}\")\n",
-        "     ),\n",
-        "    \n",
+        "    c(\n",
+        "        \"--tr-encoding\",\n",
+        "        \"-s\",\n",
+        "        \"-X POST\",\n",
+        "        glue(\"-H 'Authorization: Bearer {access_token}'\"),\n",
+        "        \"-H 'Content-Type: application/jsoin'\",\n",
+        "        url,\n",
+        "        glue(\"-d {json_instances}\"),\n",
+        "    ),\n",
         ")"
       ]
     },

notebooks/community/ml_ops/stage2/get_started_vertex_training_sklearn.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex Training for Scikit-Learn\n",
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Vertex AI Training for Scikit-Learn\n",
         "\n",
         "<table align=\"left\">\n",
         "  <td>\n",
@@ -50,9 +50,6 @@
         "    </a>\n",
         "  </td>\n",
         "</table>\n",
-        "    </a>\n",
-        "  </td>\n",
-        "</table>\n",
         "<br/><br/><br/>"
       ]
     },
@@ -68,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Training for scikit-Learn."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:custom,newsaggr,tcn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [News Aggregation](https://archive.ics.uci.edu/ml/datasets/News+Aggregator) from [ICS Machine Learning Datasets](https://archive.ics.uci.edu/ml/datasets.php). The trained model predicts the news category of the news article."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -102,6 +88,17 @@
         "- Create a `Vertex AI Model` resource."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:custom,newsaggr,tcn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [News Aggregation](https://archive.ics.uci.edu/ml/datasets/News+Aggregator) from [ICS Machine Learning Datasets](https://archive.ics.uci.edu/ml/datasets.php). The trained model predicts the news category of the news article."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -130,7 +127,7 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "**If you are using Colab or Google Cloud Notebooks**, your environment already meets\n",
+        "**If you are using Colab or Vertex AI Workbench Notebooks**, your environment already meets\n",
         "all the requirements to run this notebook. You can skip this step.\n",
         "\n",
         "**Otherwise**, make sure your environment meets this notebook's requirements.\n",
@@ -170,7 +167,7 @@
         "id": "install_mlops"
       },
       "source": [
-        "### Install additional packages\n",
+        "## Installation\n",
         "\n",
         "Install the following packages for executing this notebook."
       ]
@@ -186,7 +183,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -246,7 +243,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_vertex_training_xgboost.ipynb

@@ -66,17 +66,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Training for XGBoost."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:iris,lcn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Iris dataset](https://www.tensorflow.org/datasets/catalog/iris) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). This dataset does not require any feature engineering. The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of Iris flower species from a class of three species: setosa, virginica, or versicolor."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -97,8 +86,26 @@
         "- Training using a Python package.\n",
         "- Report accuracy when hyperparameter tuning.\n",
         "- Save the model artifacts to Cloud Storage using GCSFuse.\n",
-        "- Create a `Vertex AI Model` resource.\n",
+        "- Create a `Vertex AI Model` resource."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:iris,lcn"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is the [Iris dataset](https://www.tensorflow.org/datasets/catalog/iris) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). This dataset does not require any feature engineering. The version of the dataset in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of Iris flower species from a class of three species: setosa, virginica, or versicolor."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4fc0ad661ebb"
+      },
+      "source": [
         "### Costs \n",
         "\n",
         "This tutorial uses billable components of Google Cloud:\n",
@@ -136,7 +143,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -149,6 +156,36 @@
         "! pip3 install --upgrade google-cloud-aiplatform $USER_FLAG -q"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "oQhwq1iozAxh"
+      },
+      "source": [
+        "### Restart the kernel\n",
+        "\n",
+        "After you install the additional packages, you need to restart the notebook kernel so it can find the packages."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "zo3YFZXLzCRJ"
+      },
+      "outputs": [],
+      "source": [
+        "# Automatically restart kernel after installs\n",
+        "import os\n",
+        "\n",
+        "if not os.getenv(\"IS_TESTING\"):\n",
+        "    # Automatically restart kernel after installs\n",
+        "    import IPython\n",
+        "\n",
+        "    app = IPython.Application.instance()\n",
+        "    app.kernel.do_shutdown(True)"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -167,7 +204,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_vertex_vizier.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_vertex_vizier.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -62,18 +62,7 @@
         "## Overview\n",
         "\n",
         "\n",
-        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex Vizier."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:custom,boston,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Boston Housing Prices dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html). The version of the dataset you will use in this tutorial is built into TensorFlow. The trained model predicts the median price of a house in units of 1K USD."
+        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Vertex AI Vizier."
       ]
     },
     {
@@ -134,6 +123,38 @@
         "- multiple of objectives"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:custom,boston,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Boston Housing Prices dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html). The version of the dataset in this tutorial is built into TensorFlow. The trained model predicts the median price of a house in units of 1K USD."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "c480fc50ec3c"
+      },
+      "source": [
+        "### Costs \n",
+        "\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "* Vertex AI\n",
+        "* Cloud Storage\n",
+        "\n",
+        "\n",
+        "Learn about [Vertex AI\n",
+        "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
+        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
+        "Calculator](https://cloud.google.com/products/calculator/)\n",
+        "to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -156,7 +177,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -166,7 +187,7 @@
         "if IS_WORKBENCH_NOTEBOOK:\n",
         "    USER_FLAG = \"--user\"\n",
         "\n",
-        "! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG"
+        "! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG -q"
       ]
     },
     {
@@ -216,7 +237,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -1275,7 +1296,8 @@
         "Use the class `CustomJob` to create a custom job, such as for hyperparameter tuning, with the following parameters:\n",
         "\n",
         "- `display_name`: A human readable name for the custom job.\n",
-        "- `worker_pool_specs`: The specification for the corresponding VM instances."
+        "- `worker_pool_specs`: The specification for the corresponding VM instances.\n",
+        "- `base_output_dir`: The Cloud Storage location for storing the model artifacts."
       ]
     },
     {
@@ -1287,7 +1309,9 @@
       "outputs": [],
       "source": [
         "job = aip.CustomJob(\n",
-        "    display_name=\"boston_\" + TIMESTAMP, worker_pool_specs=worker_pool_spec\n",
+        "    display_name=\"boston_\" + TIMESTAMP,\n",
+        "    worker_pool_specs=worker_pool_spec,\n",
+        "    base_output_dir=MODEL_DIR,\n",
         ")"
       ]
     },
@@ -1420,6 +1444,32 @@
         "print(best)"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "get_best_model"
+      },
+      "source": [
+        "### Get the Best Model\n",
+        "\n",
+        "If you used the method of having the service tell the tuning script where to save the model artifacts (`DIRECT = False`), then the model artifacts for the best model are saved at:\n",
+        "\n",
+        "    MODEL_DIR/<best_trial_id>/model"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "get_best_model"
+      },
+      "outputs": [],
+      "source": [
+        "BEST_MODEL_DIR = MODEL_DIR + \"/\" + best[0] + \"/model\"\n",
+        "\n",
+        "! gsutil ls {BEST_MODEL_DIR}"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {

notebooks/community/ml_ops/stage2/get_started_with_cmek_training.ipynb

@@ -39,7 +39,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_with_cmek_training.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -64,17 +64,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with AutoML training with a customer managed encyrption key CMEK."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use in this tutorial is stored in a public #(GCS) bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip.\n"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -97,6 +86,17 @@
         "- Train an AutoML model with CMEK encryption."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset in this tutorial is stored in a public #(GCS) bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip.\n"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -139,7 +139,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -207,7 +207,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_with_logging.ipynb

@@ -0,0 +1,710 @@
+{
+  "cells": [
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "copyright"
+      },
+      "outputs": [],
+      "source": [
+        "# Copyright 2022 Google LLC\n",
+        "#\n",
+        "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
+        "# you may not use this file except in compliance with the License.\n",
+        "# You may obtain a copy of the License at\n",
+        "#\n",
+        "#     https://www.apache.org/licenses/LICENSE-2.0\n",
+        "#\n",
+        "# Unless required by applicable law or agreed to in writing, software\n",
+        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
+        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
+        "# See the License for the specific language governing permissions and\n",
+        "# limitations under the License."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "title:generic,gcp"
+      },
+      "source": [
+        "# E2E ML on GCP: MLOps stage 2 : experimentation: get started with Logging\n",
+        "\n",
+        "<table align=\"left\">\n",
+        "  <td>\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_with_logging.ipynb\">\n",
+        "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
+        "      View on GitHub\n",
+        "    </a>\n",
+        "  </td>\n",
+        "    <td>\n",
+        "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_with_logging.ipynb\">\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
+        "        </a>\n",
+        "  </td>\n",
+        "  <td>\n",
+        "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/notebooks/community/ml_ops/stage2/get_started_with_logging.ipynb\">\n",
+        "      <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
+        "      Open in Vertex AI Workbench\n",
+        "    </a>\n",
+        "  </td>\n",
+        "</table>\n",
+        "<br/><br/><br/>"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "overview:mlops"
+      },
+      "source": [
+        "## Overview\n",
+        "\n",
+        "\n",
+        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with Logging."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "objective:mlops,stage2,get_started_vertex_experiments"
+      },
+      "source": [
+        "### Objective\n",
+        "\n",
+        "In this tutorial, you learn how to use Python and Cloud logging awhen training with `Vertex AI`.\n",
+        "\n",
+        "This tutorial uses the following Google Cloud ML services:\n",
+        "\n",
+        "- `Cloud Logging`\n",
+        "\n",
+        "The steps performed include:\n",
+        "\n",
+        "- Use Python logging to log training configuration/results locally.\n",
+        "- Use Google Cloud Logging to log training configuration/results in cloud storage."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "recommendation:mlops,stage2,logging"
+      },
+      "source": [
+        "### Recommendations\n",
+        "\n",
+        "When doing E2E MLOps on Google Cloud, the following are some of the best practices for logging data when experimenting or formally training a model.\n",
+        "\n",
+        "#### Python Logging\n",
+        "\n",
+        "Use Python's logging package when doing ad-hoc training locally.\n",
+        "\n",
+        "#### Cloud Logging\n",
+        "\n",
+        "Use `Google Cloud Logging` when doing training on the cloud.\n",
+        "\n",
+        "#### Experiments\n",
+        "\n",
+        "Use Vertex AI Experiments in conjunction with logging when performing experiments to compare results for different experiment configurations."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "5341f31587c8"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "This tutorial does not use a dataset. References to example datasets is for demonstration purposes."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "41512a89f379"
+      },
+      "source": [
+        "### Costs\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "- Vertex AI\n",
+        "\n",
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "install_mlops"
+      },
+      "source": [
+        "## Installations\n",
+        "\n",
+        "Install the following packages for executing this notebook."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "install_mlops"
+      },
+      "outputs": [],
+      "source": [
+        "import os\n",
+        "\n",
+        "# The Vertex AI Workbench Notebook product has specific requirements\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
+        "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
+        "    \"/opt/deeplearning/metadata/env_version\"\n",
+        ")\n",
+        "\n",
+        "# Vertex AI Notebook requires dependencies to be installed with '--user'\n",
+        "USER_FLAG = \"\"\n",
+        "if IS_WORKBENCH_NOTEBOOK:\n",
+        "    USER_FLAG = \"--user\"\n",
+        "\n",
+        "! pip3 install --upgrade google-cloud-logging $USER_FLAG -q"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "restart"
+      },
+      "source": [
+        "### Restart the kernel\n",
+        "\n",
+        "Once you've installed the additional packages, you need to restart the notebook kernel so it can find the packages."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "restart"
+      },
+      "outputs": [],
+      "source": [
+        "import os\n",
+        "\n",
+        "if not os.getenv(\"IS_TESTING\"):\n",
+        "    # Automatically restart kernel after installs\n",
+        "    import IPython\n",
+        "\n",
+        "    app = IPython.Application.instance()\n",
+        "    app.kernel.do_shutdown(True)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "project_id"
+      },
+      "source": [
+        "## Before you begin\n",
+        "\n",
+        "### Set up your Google Cloud project\n",
+        "\n",
+        "**The following steps are required, regardless of your notebook environment.**\n",
+        "\n",
+        "1. [Select or create a Google Cloud project](https://console.cloud.google.com/cloud-resource-manager). When you first create an account, you get a $300 free credit towards your compute/storage costs.\n",
+        "\n",
+        "1. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
+        "\n",
+        "1. [Enable the Vertex AI, Compute Engine, Cloud Storage and Cloud Logging APIs](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage_component,logging).\n",
+        "\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "\n",
+        "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
+        "Cloud SDK uses the right project for all the commands in this notebook.\n",
+        "\n",
+        "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "5aee4379e8e5"
+      },
+      "source": [
+        "#### Set your project ID\n",
+        "\n",
+        "**If you don't know your project ID**, you may be able to get your project ID using `gcloud`."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "set_project_id"
+      },
+      "outputs": [],
+      "source": [
+        "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "autoset_project_id"
+      },
+      "outputs": [],
+      "source": [
+        "if PROJECT_ID == \"\" or PROJECT_ID is None or PROJECT_ID == \"[your-project-id]\":\n",
+        "    # Get your GCP project id from gcloud\n",
+        "    shell_output = ! gcloud config list --format 'value(core.project)' 2>/dev/null\n",
+        "    PROJECT_ID = shell_output[0]\n",
+        "    print(\"Project ID:\", PROJECT_ID)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "set_gcloud_project_id"
+      },
+      "outputs": [],
+      "source": [
+        "! gcloud config set project $PROJECT_ID"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "region"
+      },
+      "source": [
+        "#### Region\n",
+        "\n",
+        "You can also change the `REGION` variable, which is used for operations\n",
+        "throughout the rest of this notebook.  Below are regions supported for Vertex AI. We recommend that you choose the region closest to you.\n",
+        "\n",
+        "- Americas: `us-central1`\n",
+        "- Europe: `europe-west4`\n",
+        "- Asia Pacific: `asia-east1`\n",
+        "\n",
+        "You may not use a multi-regional bucket for training with Vertex AI. Not all regions provide support for all Vertex AI services.\n",
+        "\n",
+        "Learn more about [Vertex AI regions](https://cloud.google.com/vertex-ai/docs/general/locations)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "region"
+      },
+      "outputs": [],
+      "source": [
+        "REGION = \"[your-region]\"  # @param {type: \"string\"}\n",
+        "\n",
+        "if REGION == \"[your-region]\":\n",
+        "    REGION = \"us-central1\""
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "timestamp"
+      },
+      "source": [
+        "#### Timestamp\n",
+        "\n",
+        "If you are in a live tutorial session, you might be using a shared test account or project. To avoid name collisions between users on resources created, you create a timestamp for each instance session, and append the timestamp onto the name of resources you create in this tutorial."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "timestamp"
+      },
+      "outputs": [],
+      "source": [
+        "from datetime import datetime\n",
+        "\n",
+        "TIMESTAMP = datetime.now().strftime(\"%Y%m%d%H%M%S\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "f3bd8c0d0469"
+      },
+      "source": [
+        "### Authenticate your Google Cloud account\n",
+        "\n",
+        "**If you are using Vertex AI Workbench Notebooks**, your environment is already authenticated. Skip this step.\n",
+        "\n",
+        "**If you are using Colab**, run the cell below and follow the instructions when prompted to authenticate your account via oAuth.\n",
+        "\n",
+        "**Otherwise**, follow these steps:\n",
+        "\n",
+        "In the Cloud Console, go to the [Create service account key](https://console.cloud.google.com/apis/credentials/serviceaccountkey) page.\n",
+        "\n",
+        "1. **Click Create service account**.\n",
+        "\n",
+        "2. In the **Service account name** field, enter a name, and click **Create**.\n",
+        "\n",
+        "3. In the **Grant this service account access to project** section, click the Role drop-down list. Type \"Vertex AI\" into the filter box, and select **Vertex AI Administrator**. Type \"Storage Object Admin\" into the filter box, and select **Storage Object Admin**.\n",
+        "\n",
+        "4. Click Create. A JSON file that contains your key downloads to your local environment.\n",
+        "\n",
+        "5. Enter the path to your service account key as the GOOGLE_APPLICATION_CREDENTIALS variable in the cell below and run the cell."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "e0953a00668e"
+      },
+      "outputs": [],
+      "source": [
+        "# If you are running this notebook in Colab, run this cell and follow the\n",
+        "# instructions to authenticate your GCP account. This provides access to your\n",
+        "# Cloud Storage bucket and lets you submit training jobs and prediction\n",
+        "# requests.\n",
+        "\n",
+        "import os\n",
+        "import sys\n",
+        "\n",
+        "# If on Vertex AI Workbench, then don't execute this code\n",
+        "IS_COLAB = \"google.colab\" in sys.modules\n",
+        "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
+        "    \"DL_ANACONDA_HOME\"\n",
+        "):\n",
+        "    if \"google.colab\" in sys.modules:\n",
+        "        from google.colab import auth as google_auth\n",
+        "\n",
+        "        google_auth.authenticate_user()\n",
+        "\n",
+        "    # If you are running this notebook locally, replace the string below with the\n",
+        "    # path to your service account key and run this cell to authenticate your GCP\n",
+        "    # account.\n",
+        "    elif not os.getenv(\"IS_TESTING\"):\n",
+        "        %env GOOGLE_APPLICATION_CREDENTIALS ''"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "setup_vars"
+      },
+      "source": [
+        "### Set up variables\n",
+        "\n",
+        "Next, set up some variables used throughout the tutorial.\n",
+        "### Import libraries"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "import_aip:mbsdk"
+      },
+      "outputs": [],
+      "source": [
+        "import logging\n",
+        "\n",
+        "import google.cloud.aiplatform as aiplatform"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "init_aip:mbsdk,region"
+      },
+      "source": [
+        "### Initialize Vertex AI SDK for Python\n",
+        "\n",
+        "Initialize the Vertex AI SDK for Python for your project and corresponding bucket."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "init_aip:mbsdk,region"
+      },
+      "outputs": [],
+      "source": [
+        "aiplatform.init(project=PROJECT_ID, location=REGION)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "python_logging"
+      },
+      "source": [
+        "## Python Logging\n",
+        "\n",
+        "The Python logging package is widely used for logging within Python scripts. Commonly used features:\n",
+        "\n",
+        "- Set logging levels.\n",
+        "- Send log output to console.\n",
+        "- Send log output to a file.\n",
+        "\n",
+        "### Logging Levels in Python Logging\n",
+        "\n",
+        "The logging levels in order (from least to highest) and each level inclusive of the previous level are :\n",
+        "\n",
+        "1. Informational\n",
+        "2. Warnings\n",
+        "3. Errors\n",
+        "4. Debugging\n",
+        "\n",
+        "By default, the logging level is set to error level.\n",
+        "\n",
+        "### Logging output to console\n",
+        "\n",
+        "By default, the Python logging package outputs to the console. Note, in the example the debug log message is not outputted since the default logging level is set to error."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "python_logging"
+      },
+      "outputs": [],
+      "source": [
+        "def logging_examples():\n",
+        "    logging.info(\"Model training started...\")\n",
+        "    logging.warning(\"Using older version of package ...\")\n",
+        "    logging.error(\"Training was terminated ...\")\n",
+        "    logging.debug(\"Hyperparameters were ...\")\n",
+        "\n",
+        "\n",
+        "logging_examples()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "python_logging_level"
+      },
+      "source": [
+        "### Setting logging level\n",
+        "\n",
+        "To set the logging level, you get the logging handler using `getLogger()`. You can have multiple logging handles. When `getLogger()` is called without any arguments, it gets the default handler named ROOT. With the handler, you set the logging level with the method `setLevel()`."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "python_logging_level"
+      },
+      "outputs": [],
+      "source": [
+        "logging.getLogger().setLevel(logging.DEBUG)\n",
+        "\n",
+        "logging_examples()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "python_logging_remove"
+      },
+      "source": [
+        "### Clearing handlers\n",
+        "\n",
+        "At times, you may desire to reconfigure your logging. A common practice in this case is to first remove all existing logging handles for a fresh start."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "python_logging_remove"
+      },
+      "outputs": [],
+      "source": [
+        "for handler in logging.root.handlers[:]:\n",
+        "    logging.root.removeHandler(handler)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "python_logging_file"
+      },
+      "source": [
+        "### Output to a local file\n",
+        "\n",
+        "You can preserve your logging output to a file that is local to where the Python script is running with the method `BasicConfig()`, that takes the following parameters:\n",
+        "\n",
+        "- `filename`: The file path to the local file to write the log output to.\n",
+        "- `level`: Sets the level of logging that is written to the logging file.\n",
+        "\n",
+        "*Note:* You cannot use a Cloud Storage bucket as the output file."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "python_logging_file"
+      },
+      "outputs": [],
+      "source": [
+        "logging.basicConfig(filename=\"mylog.log\", level=logging.DEBUG)\n",
+        "\n",
+        "logging_examples()\n",
+        "\n",
+        "! cat mylog.log"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "cloud_logging"
+      },
+      "source": [
+        "## Logging with Google Cloud Logging\n",
+        "\n",
+        "You can preserve and retrieve your logging output to `Google Cloud Logging` service. Commonly used features:\n",
+        "\n",
+        "- Set logging levels.\n",
+        "- Send log output to storage.\n",
+        "- Retrieve log output from storage.\n",
+        "\n",
+        "### Logging Levels in Cloud Logging\n",
+        "\n",
+        "The logging levels in order (from least to highest) are, with each level inclusive of the previous level:\n",
+        "\n",
+        "1. Informational\n",
+        "2. Warnings\n",
+        "3. Errors\n",
+        "4. Debugging\n",
+        "\n",
+        "By default, the logging level is set to warning level.\n",
+        "\n",
+        "### Configurable and storing log data.\n",
+        "\n",
+        "To use the `Google Cloud Logging` service, you do the following steps:\n",
+        "\n",
+        "1. Create a client to the service.\n",
+        "2. Obtain a handler for the service.\n",
+        "3. Create a logger instance and set logging level.\n",
+        "4. Attach logger instance to the service.\n",
+        "\n",
+        "Learn more about [Logging client libraries](https://cloud.google.com/logging/docs/reference/libraries)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "cloud_logging"
+      },
+      "outputs": [],
+      "source": [
+        "import google.cloud.logging\n",
+        "from google.cloud.logging.handlers import CloudLoggingHandler\n",
+        "\n",
+        "# Connect to the Cloud Logging service\n",
+        "cl_client = google.cloud.logging.Client(project=PROJECT_ID)\n",
+        "handler = CloudLoggingHandler(cl_client, name=\"mylog\")\n",
+        "\n",
+        "# Create a logger instance and logging level\n",
+        "cloud_logger = logging.getLogger(\"cloudLogger\")\n",
+        "cloud_logger.setLevel(logging.INFO)\n",
+        "\n",
+        "# Attach the logger instance to the service.\n",
+        "cloud_logger.addHandler(handler)\n",
+        "\n",
+        "# Log something\n",
+        "cloud_logger.error(\"bad news\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "cloud_logging_write"
+      },
+      "source": [
+        "### Logging output\n",
+        "\n",
+        "Logging output at specific levels is identical to Python logging with respect to method and method names. The only difference is that you use your instance of the cloud logger in place of logging."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "cloud_logging_write"
+      },
+      "outputs": [],
+      "source": [
+        "cloud_logger.info(\"Model training started...\")\n",
+        "cloud_logger.warning(\"Using older version of package ...\")\n",
+        "cloud_logger.error(\"Training was terminated ...\")\n",
+        "cloud_logger.debug(\"Hyperparameters were ...\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "cloud_logging_list"
+      },
+      "source": [
+        "### Get logging entries\n",
+        "\n",
+        "To get the logged output, you:\n",
+        "\n",
+        "1. Retrieve the log handle to the service.\n",
+        "2. Using the handle, call the method `list_entries()`.\n",
+        "3. Iterate through the entries."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "cloud_logging_list"
+      },
+      "outputs": [],
+      "source": [
+        "logger = cl_client.logger(\"mylog\")\n",
+        "\n",
+        "for entry in logger.list_entries():\n",
+        "    timestamp = entry.timestamp.isoformat()\n",
+        "    print(\"* {}: {}: {}\".format(timestamp, entry.severity, entry.payload))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "cleanup:mbsdk"
+      },
+      "source": [
+        "# Cleaning up\n",
+        "\n",
+        "To clean up all Google Cloud resources used in this project, you can [delete the Google Cloud\n",
+        "project](https://cloud.google.com/resource-manager/docs/creating-managing-projects#shutting_down_projects) you used for the tutorial.\n",
+        "\n",
+        "Otherwise, you can delete the individual resources you created in this tutorial."
+      ]
+    }
+  ],
+  "metadata": {
+    "colab": {
+      "name": "get_started_with_logging.ipynb",
+      "toc_visible": true
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "name": "python3"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}

notebooks/community/ml_ops/stage2/get_started_with_tabnet.ipynb

@@ -39,7 +39,7 @@
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samplestree/main/notebooks/community/ml_ops/stage2/get_started_with_tabnet.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_with_tabnet.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
@@ -67,15 +67,18 @@
         "\n",
         "TabNet uses a machine learning technique called sequential attention to select which model features to reason from at each step in the model. This mechanism makes it possible to explain how the model arrives at its predictions and helps it learn more accurate models. TabNet not only outperforms other neural networks and decision trees but also provides interpretable feature attributions. \n",
         "\n",
-        "Research paper: [TabNet: Attentive Interpretable Tabular Learning](https://arxiv.org/pdf/1908.07442.pdf)\n",
-        "\n",
-        "### Dataset\n",
-        "\n",
-        "This tutorial uses the `petfinder` in the public Cloud Storage bucket `gs://cloud-samples-data/ai-platform-unified/datasets/tabular/`, which was generated from the [PetFinder.my Adoption Prediction](https://www.kaggle.com/c/petfinder-adoption-prediction). This dataset predicts how quickly an animal will be adopted.\n",
-        "\n",
+        "Research paper: [TabNet: Attentive Interpretable Tabular Learning](https://arxiv.org/pdf/1908.07442.pdf)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "c5040751873a"
+      },
+      "source": [
         "### Objective\n",
         "\n",
-        "In this notebook, you will learn how to run `Vertex AI TabNet` built algorithm for training custom tabular models.\n",
+        "In this notebook, you learn how to run `Vertex AI TabNet` built algorithm for training custom tabular models.\n",
         "\n",
         "This tutorial uses the following Google Cloud ML services and resources:\n",
         "\n",
@@ -93,11 +96,28 @@
         "- Deploy the `Vertex AI Model` resource to a `Vertex AI Endpoint` resource.\n",
         "- Make a prediction with the deployed model.\n",
         "- Hyperparameter tuning the `Vertex AI TabNet` model.\n",
-        "- Train the model using `Vertex AI Training` using BigQuery table.\n",
+        "- Train the model using `Vertex AI Training` using BigQuery table."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ac8c8586ab03"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "This tutorial uses the `petfinder` in the public Cloud Storage bucket `gs://cloud-samples-data/ai-platform-unified/datasets/tabular/`, which was generated from the [PetFinder.my Adoption Prediction](https://www.kaggle.com/c/petfinder-adoption-prediction). This dataset predicts how quickly an animal is adopted.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4fc0ad661ebb"
+      },
+      "source": [
         "### Costs \n",
         "\n",
-        "\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
         "* Vertex AI\n",
@@ -133,7 +153,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -143,8 +163,8 @@
         "if IS_WORKBENCH_NOTEBOOK:\n",
         "    USER_FLAG = \"--user\"\n",
         "\n",
-        "! pip3 install {USER_FLAG} --upgrade tensorflow\n",
-        "! pip3 install {USER_FLAG} --upgrade google-cloud-aiplatform tensorboard-plugin-profile\n",
+        "! pip3 install {USER_FLAG} --upgrade tensorflow -q\n",
+        "! pip3 install {USER_FLAG} --upgrade google-cloud-aiplatform tensorboard-plugin-profile -q\n",
         "! gcloud components update --quiet"
       ]
     },
@@ -200,7 +220,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_with_tfhub_models.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/get_started_with_tfhub_models.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation: get started with prebuilt TensorFlow Hub (TFHub) models."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [CIFAR10 dataset](https://www.tensorflow.org/datasets/catalog/cifar10) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use is built into TensorFlow. The trained model predicts which type of class an image is from ten classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -103,6 +92,17 @@
         "    - Save model artifacts and upload as Vertex AI Model resource."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [CIFAR10 dataset](https://www.tensorflow.org/datasets/catalog/cifar10) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset is built into TensorFlow. The trained model predicts which type of class an image is from ten classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -145,7 +145,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -157,27 +157,8 @@
         "    USER_FLAG = \"\"\n",
         "\n",
         "! pip3 install --upgrade google-cloud-aiplatform $USER_FLAG -q\n",
-        "! pip3 install tensorflow-datasets $USER_FLAG -q"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "install_storage"
-      },
-      "source": [
-        "Install the latest GA version of *google-cloud-storage* library as well."
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "install_storage"
-      },
-      "outputs": [],
-      "source": [
-        "! pip3 install -U google-cloud-storage $USER_FLAG"
+        "! pip3 install tensorflow-datasets $USER_FLAG -q\n",
+        "! pip3 install -U google-cloud-storage $USER_FLAG -q"
       ]
     },
     {
@@ -231,7 +212,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage2/get_started_with_visionapi_and_automl.ipynb

@@ -29,7 +29,7 @@
         "id": "3f8c2f702ccd"
       },
       "source": [
-        "This is an updated version of a notebook contributed by [Mohammad Al-Ansari](https://github.com/Mansari). Special thanks to [Andrew Ferlitsch](https://github.com/andrewferlitsch) for his reviews and edits.\n",
+        "This notebook is an updated version of a notebook contributed by [Mohammad Al-Ansari](https://github.com/Mansari). Special thanks to [Andrew Ferlitsch](https://github.com/andrewferlitsch) for his reviews and edits.\n",
         "\n",
         "This is an extension of the [Vertex AI SDK for Python: AutoML training text entity extraction model for online prediction notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_text_entity_extraction_online.ipynb) originally co-authored by [Andrew Ferlitsch](https://github.com/andrewferlitsch) and [\n",
         "Karl Weinmeister](https://github.com/kweinmeister). This version add the use of `Vision API` and `BigQuery` to preprocess a `Vertex AI AutoML` dataset for text entity extraction model training."
@@ -76,21 +76,6 @@
         "This tutorial demonstrates how to use `BigQuery`, `Vision AI`, and `Vertex AI SDK` for Python to train a text entity extraction model based on existing training data."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:biomedical,ten"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Patent PDF Samples with Extracted Structured Data](https://console.cloud.google.com/marketplace/product/global-patents/labeled-patents) from Google Public Data Sets. \n",
-        "\n",
-        "This dataset includes data extracted from over 300 patent documents issued in the US and EU. The dataset includes links to Google Cloud Storage blobs for the first page of each patent, in addition to a number of extracted entities. \n",
-        "\n",
-        "The data is published as a [public dataset](https://cloud.google.com/bigquery/public-data) on `BigQuery`."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -99,7 +84,7 @@
       "source": [
         "### Objective\n",
         "\n",
-        "In this tutorial, you create an `AutoML` text entity extraction model pre-existing extracted data by generating a custom import file. You will deploy this mode for online prediction from a Python script using the `BigQuery`, `Vision AI`, Cloud Storage and `Vertex AI SDK` for Python. You can alternatively create and deploy models using the `gcloud` command-line tool or online using the Cloud Console.\n",
+        "In this tutorial, you create an `AutoML` text entity extraction model pre-existing extracted data by generating a custom import file. You deploy this mode for online prediction from a Python script using the `BigQuery`, `Vision AI`, Cloud Storage and `Vertex AI SDK` for Python. You can alternatively create and deploy models using the `gcloud` command-line tool or online using the Cloud Console.\n",
         "\n",
         "Using existing training data that have been previously annotated can be very useful in training a model, as it allows you to use a larger data set with minimal resources.\n",
         "\n",
@@ -121,6 +106,21 @@
         "- Undeploy the `Model`."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:biomedical,ten"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Patent PDF Samples with Extracted Structured Data](https://console.cloud.google.com/marketplace/product/global-patents/labeled-patents) from Google Public Data Sets. \n",
+        "\n",
+        "This dataset includes data extracted from over 300 patent documents issued in the US and EU. The dataset includes links to Google Cloud Storage blobs for the first page of each patent, in addition to a number of extracted entities. \n",
+        "\n",
+        "The data is published as a [public dataset](https://cloud.google.com/bigquery/public-data) on `BigQuery`."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -151,7 +151,7 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "If you are using Colab or Google Cloud Notebooks, your environment already meets all the requirements to run this notebook. You can skip this step.\n",
+        "If you are using Colab or Vertex AI Workbench Notebooks, your environment already meets all the requirements to run this notebook.  \n",
         "\n",
         "Otherwise, make sure your environment meets this notebook's requirements. You need the following:\n",
         "\n",
@@ -201,7 +201,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -265,7 +265,7 @@
         "\n",
         "3. [Enable the following APIs: BigQuery APIs, Vision API, Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=bigquery.googleapis.com,vision.googleapis.com,aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -273,6 +273,17 @@
         "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$`."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "WReHDGG5g0XY"
+      },
+      "source": [
+        "#### Set your project ID\n",
+        "\n",
+        "**If you don't know your project ID**, you may be able to get your project ID using `gcloud`."
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,

notebooks/community/ml_ops/stage2/mlops_experimentation.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "  <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage2/mlops_experimentation.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 2 : experimentation."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:bq,chicago,lbn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Chicago Taxi](https://www.kaggle.com/chicago/chicago-taxi-trips-bq). The version of the dataset you will use in this tutorial is stored in a public BigQuery table. The trained model predicts whether someone would leave a tip for a taxi fare."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -160,6 +149,39 @@
         "     - If greater then baseline, then upload model as the new baseline and save evaluation results with the model."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:bq,chicago,lbn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Chicago Taxi](https://www.kaggle.com/chicago/chicago-taxi-trips-bq). The version of the dataset in this tutorial is stored in a public BigQuery table. The trained model predicts whether someone leaves a tip for a taxi fare."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "costs"
+      },
+      "source": [
+        "### Costs\n",
+        "\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "* BigQuery\n",
+        "* Vision API\n",
+        "* Vertex AI\n",
+        "* Cloud Storage\n",
+        "\n",
+        "Learn about [Vertex AI\n",
+        "pricing](https://cloud.google.com/vertex-ai/pricing), [Vision API pricing](https://cloud.google.com/vision/pricing), [BigQuery pricing](https://cloud.google.com/bigquery/pricing), [Cloud Storage\n",
+        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
+        "Calculator](https://cloud.google.com/products/calculator/)\n",
+        "to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -182,7 +204,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -194,20 +216,20 @@
         "\n",
         "ONCE_ONLY = False\n",
         "if ONCE_ONLY:\n",
-        "    ! pip3 install -U tensorflow==2.5 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-data-validation==1.2 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-transform==1.2 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-io==0.18 $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-pipeline-components $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-bigquery $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-logging $USER_FLAG\n",
-        "    ! pip3 install --upgrade apache-beam[gcp] $USER_FLAG\n",
-        "    ! pip3 install --upgrade pyarrow $USER_FLAG\n",
-        "    ! pip3 install --upgrade cloudml-hypertune $USER_FLAG\n",
-        "    ! pip3 install --upgrade kfp $USER_FLAG\n",
-        "    ! pip3 install --upgrade torchvision $USER_FLAG\n",
-        "    ! pip3 install --upgrade rpy2 $USER_FLAG"
+        "    ! pip3 install -U tensorflow==2.5 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-data-validation==1.2 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-transform==1.2 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-io==0.18 $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-pipeline-components $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-bigquery $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-logging $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade apache-beam[gcp] $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade pyarrow $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade cloudml-hypertune $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade kfp $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade torchvision $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade rpy2 $USER_FLAG -q"
       ]
     },
     {
@@ -257,7 +279,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com). \n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage3/README.md

@@ -33,79 +33,60 @@ The third stage in MLOps is formalization to develop an automated pipeline proce
 
 ### Get Started
 
-[Get Started with Kubeflow pipelines](get_started_with_kubeflow_pipelines.ipynb)
+[Get started with AutoML Tabular Pipeline Workflows](get_started_with_automl_tabular_pipeline_workflow.ipynb)
 
 ```
 The steps performed include:
+- Define training specification.
+    - Dataset specification
+    - Hyperparameter overide specification
+    - machine specifications
+- Construct tabular workflow pipeline.
+- Compile and execute pipeline.
+- View evaluation metrics artifact.
+- Export AutoML model as an OSS TF model.
+- Create `Endpoint` resource.
+- Deploy exported OSS TF model.
+- Make a prediction.
 
-- Building KFP lightweight Python function components.
-- Assembling and compiling KFP components into a pipeline.
-- Executing a KFP pipeline using Vertex AI Pipelines.
-- Loading component and pipeline definitions from a source code repository.
-- Building sequential, parallel, multiple output components.
-- Building control flow into pipelines.
 ```
 
-[Get Started with BQ and TFDV components](get_started_with_bq_tfdv_pipeline_components.ipynb)
+[Get started with Vertex AI Model Registry](get_started_with_model_registry.ipynb)
 
 ```
 The steps performed include:
-
-- Build and execute a pipeline component for creating a Vertex AI Tabular Dataset from a BigQuery table.
-- Build and execute a pipeline component for generating TFDV statistics and schema from a Vertex AI Tabular Dataset.
-- Execute a Vertex AI pipeline.
+- Create and register a first version of a model to `Vertex AI Model Registry`.
+- Create and register a second version of a model to `Vertex AI Model Registry`.
+- Updating the model version which is the default (blessed).
+- Deleting a model version.
+- Retraining the next model version.
 ```
 
-[Get Started with Dataflow components](get_started_with_dataflow_pipeline_components.ipynb)
+[Get started with Dataproc serverless pipeline components](get_started_with_dataproc_serverless_pipeline_components.ipynb)
 
 ```
 The steps performed include:
+- `DataprocPySparkBatchOp` for running PySpark batch workloads.
+- `DataprocSparkBatchOp` for running Spark batch workloads.
+- `DataprocSparkSqlBatchOp` for running Spark SQL batch workloads.
+- `DataprocSparkRBatchOp` for running SparkR batch workloads.
 
-- Build an Apache Beam data pipeline.
-- Encapsulate the Apache Beam data pipeline with a Dataflow component in a Vertex AI pipeline.
-- Execute a Vertex AI pipeline.
 ```
 
-[Get Started with Dataproc components](get_started_with_dataproc_pipeline_components.ipynb)
+[Get started with TFX pipelines](get_started_with_tfx_pipeline.ipynb)
 
 ```
 The steps performed include:
-
-- DataprocPySparkBatchOp for PySpark batch workloads.
-- DataprocSparkBatchOp for Spark batch workloads.
-- DataprocSparkSqlBatchOp for running Spark SQL batch workloads.
-- DataprocSparkRBatchOp for running SparkR batch workloads.
+- Create a TFX e2e pipeline.
+- Execute the pipeline locally.
+- Execute the pipeline on Google Cloud using `Vertex AI Training`
+- Execute the pipeline using `Vertex AI Pipelines`.
 ```
 
-[Get Started with Vertex AI AutoML components](get_started_with_automl_pipeline_components.ipynb)
+[Get started with Vertex AI Hyperparameter Tuning pipeline components](get_started_with_hpt_pipeline_components.ipynb)
 
 ```
 The steps performed include:
-
-- Construct a pipeline for:
-    - Training a Vertex AI AutoML trained model.
-    - Test the serving binary with a batch prediction job.
-    - Deploying a Vertex AI AutoML trained model.
-- Execute a Vertex AI pipeline.
-```
-
-[Get Started with Vertex AI Custom Training components](get_started_with_custom_training_pipeline_components.ipynb)
-
-```
-The steps performed include:
-
-- Construct a pipeline for:
-    - Training a Vertex AI custom trained model.
-    - Test the serving binary with a batch prediction job.
-    - Deploying a Vertex AI custom trained model.
-- Execute a Vertex AI pipeline.
-```
-
-[Get Started with Vertex AI Hyperparameter Tuning components](get_started_with_hpt_pipeline_components.ipynb)
-
-```
-The steps performed include:
-
 - Construct a pipeline for:
     - Hyperparameter tune/train a custom model.
     - Retrieve the tuned hyperparameter values and metrics to optimize.
@@ -113,13 +94,95 @@ The steps performed include:
       - Get the location of the model artifacts for the best tuned model.
       - Upload the model artifacts to a `Vertex AI Model` resource.
 - Execute a Vertex AI pipeline.
+
 ```
 
-[Get Started with BQML components](get_started_with_bqml_pipeline_components.ipynb)
+[Get started with Apache Airflow and Vertex AI Pipelines](get_started_with_airflow_and_vertex_pipelines.ipynb)
 
 ```
 The steps performed include:
+- Create Cloud Composer environment.
+- Upload Airflow DAG to Composer environment that performs data processing -- i.e., creates a BigQuery table from a CSV file.
+- Create a `Vertex AI Pipeline` that triggers the Airflow DAG.
+- Execute the `Vertex AI Pipeline`.
 
+```
+
+[Get started with Vertex AI custom training pipeline components](get_started_with_custom_training_pipeline_components.ipynb)
+
+```
+The steps performed include:
+- Construct a pipeline for:
+    - Training a Vertex AI custom trained model.
+    - Test the serving binary with a batch prediction job.
+    - Deploying a Vertex AI custom trained model.
+- Execute a Vertex AI pipeline.
+- Construct a pipeline for:
+     - Construct a custom training component.
+     - Convert custom training component to CustomTrainingJobOp.
+     - Training a Vertex AI custom trained model using the converted component.
+    - Deploying a Vertex AI custom trained model.
+- Execute a Vertex AI pipeline.
+```
+
+[Get started with AutoML pipeline components](get_started_with_automl_pipeline_components.ipynb)
+
+```
+The steps performed include:
+- Construct a pipeline for:
+    - Training a Vertex AI AutoML trained model.
+    - Test the serving binary with a batch prediction job.
+    - Deploying a Vertex AI AutoML trained model.
+- Execute a Vertex AI pipeline.
+
+```
+
+[Get started with Kubeflow pipelines](get_started_with_kubeflow_pipelines.ipynb)
+
+```
+The steps performed include:
+- Building KFP lightweight Python function components.
+- Assembling and compiling KFP components into a pipeline.
+- Executing a KFP pipeline using Vertex AI Pipelines.
+- Loading component and pipeline definitions from a source code repository.
+- Building sequential, parallel, multiple output components.
+- Building control flow into pipelines.
+
+```
+
+[Get started with machine management for Vertex AI Pipelines](get_started_with_machine_management.ipynb)
+
+```
+The steps performed in this tutorial include:
+- Create a custom component with a self-contained training job.
+- Execute pipeline using component-level settings for machine resources
+- Convert the self-contained training component into a `Vertex AI CustomJob`.
+- Execute pipeline using customjob-level settings for machine resources 
+
+```
+
+[Get started with BigQuery and TFDV pipeline components](get_started_with_bq_tfdv_pipeline_components.ipynb)
+
+```
+The steps performed include:
+- Build and execute a pipeline component for creating a Vertex AI Tabular Dataset from a BigQuery table.
+- Build and execute a pipeline component for generating TFDV statistics and schema from a Vertex AI Tabular Dataset.
+- Execute a Vertex AI pipeline.
+```
+
+[Get started with Dataflow pipeline components](get_started_with_dataflow_pipeline_components.ipynb)
+
+```
+The steps performed include:
+- Build an Apache Beam data pipeline.
+- Encapsulate the Apache Beam data pipeline with a Dataflow component in a Vertex AI pipeline.
+- Execute a Vertex AI pipeline.
+```
+
+[Get started with BigQuery ML pipeline components](get_started_with_bqml_pipeline_components.ipynb)
+
+```
+The steps performed include:
 - Construct a pipeline for:
     - Training BigQuery ML model.
     - Evaluating the BigQuery ML model.
@@ -130,12 +193,10 @@ The steps performed include:
 - Make a prediction with the deployed Vertex AI model.
 ```
 
-[Get Started with rapid prototyping with BQML and AutoML components](get_started_with_rapid_prototyping_bqml_automl.ipynb)
-
+[Get started with rapid prototyping with AutoML and BigQuery ML](get_started_with_rapid_prototyping_bqml_automl.ipynb)
 
 ```
 The steps performed include:
-
 - Creating a BigQuery and Vertex AI training dataset.
 - Training a BigQuery ML and AutoML model.
 - Extracting evaluation metrics from the BigQueryML and AutoML models.
@@ -144,38 +205,6 @@ The steps performed include:
 - Testing the deployed model infrastructure.
 ```
 
-[Get Started with TFX Pipelines with Vertex AI](get_started_with_tfx_pipeline.ipynb)
-
-```
-The steps performed include:
-
-- Create a TFX e2e pipeline.
-- Execute the pipeline locally.
-- Execute the pipeline on Google Cloud using `Vertex AI Training`
-- Execute the pipeline using `Vertex AI Pipelines`.
-```
-
-[Get Started with machine management](get_started_with_machine_management.ipynb)
-
-```
-The steps performed in this tutorial include:
-
-- Create a custom component with a self-contained training job.
-- Execute pipeline using component-level settings for machine resources
-- Convert the self-contained training componnt into a Vertex AI CustomJob.
-- Execute pipeline using customjob-level settings for machine resources
-```
-
-[Get Started with Apache Airflow and Vertex AI Pipelines](get_started_with_airflow_and_vertex_pipelines.ipynb)
-
-```
-The steps performed in this tutorial include:
-
-- Create Cloud Composer environment.
-- Upload Airflow DAG to Composer environment that performs data processing -- i.e., creates a BigQuery table from a CSV file.
-- Create a Vertex Pipeline that triggers the Airflow DAG.
-- Execute the `Vertex AI Pipeline`.
-```
 
 ### E2E Stage Example
 
@@ -183,7 +212,6 @@ The steps performed in this tutorial include:
 
 ```
 The steps performed include:
-
 - Obtain resources from the experimentation stage.
     - Baseline model.
     - Dataset schema/statistics for baseline model.
@@ -196,3 +224,4 @@ The steps performed include:
     - Create the Vertex AI Model base model.
 - Formalize a training pipeline.
 ```
+

notebooks/community/ml_ops/stage3/get_started_with_airflow_and_vertex_pipelines.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "    <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage3/get_started_with_airflow_and_vertex_pipelines.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -64,17 +64,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with Apache Airflow and Vertex AI Pipelines."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is [Condensed Game Data](gs://example-datasets/game_data_condensed.csv), which comes from the [Apache Beam examples](https://github.com/apache/beam/tree/master/sdks/python/apache_beam/examples/complete/game). The version used in this tutorial is stored in a Cloud Storage bucket."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -97,16 +86,35 @@
         "- Create Cloud Composer environment.\n",
         "- Upload Airflow DAG to Composer environment that performs data processing -- i.e., creates a BigQuery table from a CSV file.\n",
         "- Create a `Vertex AI Pipeline` that triggers the Airflow DAG.\n",
-        "- Execute the `Vertex AI Pipeline`.\n",
+        "- Execute the `Vertex AI Pipeline`."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is [Condensed Game Data](gs://example-datasets/game_data_condensed.csv), which comes from the [Apache Beam examples](https://github.com/apache/beam/tree/master/sdks/python/apache_beam/examples/complete/game). The version used in this tutorial is stored in a Cloud Storage bucket."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "b8a374d1a7dc"
+      },
+      "source": [
         "### Costs\n",
         "\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
         "- Vertex AI\n",
         "- Cloud Storage\n",
+        "- BigQuery\n",
         "\n",
-        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and [BigQuery pricing](https://cloud.google.com/bigquery/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
       ]
     },
     {
@@ -131,7 +139,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -182,6 +190,8 @@
         "id": "ce9e86b26403"
       },
       "source": [
+        "#### Check package versions\n",
+        "\n",
         "Check that you have correctly installed the packages. The KFP SDK version should be >=1.6:"
       ]
     },
@@ -203,6 +213,8 @@
         "id": "BF1j6f9HApxa"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -213,7 +225,7 @@
         "\n",
         "1. [Enable the Vertex AI](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com) and [Composer API](https://console.cloud.google.com/flows/enableapi?apiid=composer.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -747,7 +759,7 @@
       "source": [
         "# This code is modified version of https://github.com/GoogleCloudPlatform/python-docs-samples/blob/master/composer/rest/get_client_id.py\n",
         "\n",
-        "shell_output=! python3 get_composer_config.py $PROJECT_ID $REGION $COMPOSER_ENV_NAME\n",
+        "shell_output = ! python3 get_composer_config.py $PROJECT_ID $REGION $COMPOSER_ENV_NAME\n",
         "COMPOSER_WEB_URI = shell_output[0]\n",
         "COMPOSER_DAG_GCS = shell_output[1]\n",
         "COMPOSER_CLIENT_ID = shell_output[2]\n",
@@ -977,7 +989,7 @@
         "    dag_name: str,\n",
         "    composer_client_id: str,\n",
         "    composer_webserver_id: str,\n",
-        "    response: Output[Artifact]\n",
+        "    response: Output[Artifact],\n",
         "):\n",
         "    # [START composer_trigger]\n",
         "\n",
@@ -988,10 +1000,9 @@
         "    from google.auth.transport.requests import Request\n",
         "    from google.oauth2 import id_token\n",
         "\n",
+        "    IAM_SCOPE = \"https://www.googleapis.com/auth/iam\"\n",
+        "    OAUTH_TOKEN_URI = \"https://www.googleapis.com/oauth2/v4/token\"\n",
         "\n",
-        "    IAM_SCOPE = 'https://www.googleapis.com/auth/iam'\n",
-        "    OAUTH_TOKEN_URI = 'https://www.googleapis.com/oauth2/v4/token'\n",
-        "    \n",
         "    data = '{\"replace_microseconds\":\"false\"}'\n",
         "    context = None\n",
         "\n",
@@ -1008,13 +1019,13 @@
         "    \"\"\"\n",
         "\n",
         "    # Form webserver URL to make REST API calls\n",
-        "    webserver_url = f'{composer_webserver_id}/api/experimental/dags/{dag_name}/dag_runs'\n",
+        "    webserver_url = f\"{composer_webserver_id}/api/experimental/dags/{dag_name}/dag_runs\"\n",
         "    # print(webserver_url)\n",
         "\n",
         "    # This code is copied from\n",
         "    # https://github.com/GoogleCloudPlatform/python-docs-samples/blob/master/iap/make_iap_request.py\n",
         "    # START COPIED IAP CODE\n",
-        "    def make_iap_request(url, client_id, method='GET', **kwargs):\n",
+        "    def make_iap_request(url, client_id, method=\"GET\", **kwargs):\n",
         "        \"\"\"Makes a request to an application protected by Identity-Aware Proxy.\n",
         "        Args:\n",
         "          url: The Identity-Aware Proxy-protected URL to fetch.\n",
@@ -1028,8 +1039,8 @@
         "          The page body, or raises an exception if the page couldn't be retrieved.\n",
         "        \"\"\"\n",
         "        # Set the default timeout, if missing\n",
-        "        if 'timeout' not in kwargs:\n",
-        "            kwargs['timeout'] = 90\n",
+        "        if \"timeout\" not in kwargs:\n",
+        "            kwargs[\"timeout\"] = 90\n",
         "\n",
         "        # Obtain an OpenID Connect (OIDC) token from metadata server or using service\n",
         "        # account.\n",
@@ -1039,32 +1050,41 @@
         "        # Authorization header containing \"Bearer \" followed by a\n",
         "        # Google-issued OpenID Connect token for the service account.\n",
         "        resp = requests.request(\n",
-        "            method, url,\n",
-        "            headers={'Authorization': 'Bearer {}'.format(\n",
-        "                google_open_id_connect_token)}, **kwargs)\n",
+        "            method,\n",
+        "            url,\n",
+        "            headers={\"Authorization\": \"Bearer {}\".format(google_open_id_connect_token)},\n",
+        "            **kwargs,\n",
+        "        )\n",
         "        if resp.status_code == 403:\n",
-        "            raise Exception('Service account does not have permission to '\n",
-        "                            'access the IAP-protected application.')\n",
+        "            raise Exception(\n",
+        "                \"Service account does not have permission to \"\n",
+        "                \"access the IAP-protected application.\"\n",
+        "            )\n",
         "        elif resp.status_code != 200:\n",
         "            raise Exception(\n",
-        "                'Bad response from application: {!r} / {!r} / {!r}'.format(\n",
-        "                    resp.status_code, resp.headers, resp.text))\n",
+        "                \"Bad response from application: {!r} / {!r} / {!r}\".format(\n",
+        "                    resp.status_code, resp.headers, resp.text\n",
+        "                )\n",
+        "            )\n",
         "        else:\n",
         "            print(f\"response = {resp.text}\")\n",
         "            # not executed when testing locally\n",
         "            if response:\n",
         "                file_path = os.path.join(response.path)\n",
         "                os.makedirs(file_path)\n",
-        "                with open(os.path.join(file_path, \"airflow_response.json\"), 'w') as f:\n",
+        "                with open(os.path.join(file_path, \"airflow_response.json\"), \"w\") as f:\n",
         "                    json.dump(resp.text, f)\n",
         "\n",
         "    # END COPIED IAP CODE\n",
         "\n",
-        "    \n",
         "    # Make a POST request to IAP which then Triggers the DAG\n",
         "    make_iap_request(\n",
-        "        webserver_url, composer_client_id, method='POST', json={\"conf\": data, \"replace_microseconds\": 'false'})\n",
-        "    \n",
+        "        webserver_url,\n",
+        "        composer_client_id,\n",
+        "        method=\"POST\",\n",
+        "        json={\"conf\": data, \"replace_microseconds\": \"false\"},\n",
+        "    )\n",
+        "\n",
         "    # [END composer_trigger]"
       ]
     },
@@ -1094,7 +1114,7 @@
         "        dag_name=COMPOSER_DAG_NAME,\n",
         "        composer_client_id=COMPOSER_CLIENT_ID,\n",
         "        composer_webserver_id=COMPOSER_WEB_URI,\n",
-        "        response=None\n",
+        "        response=None,\n",
         "    )\n",
         "except Exception as e:\n",
         "    print(e)"
@@ -1121,12 +1141,13 @@
       },
       "outputs": [],
       "source": [
-        "PATH=%env PATH\n",
+        "PATH = %env PATH\n",
         "%env PATH={PATH}:/home/jupyter/.local/bin\n",
         "\n",
         "PIPELINE_ROOT = f\"{BUCKET_URI}/pipeline_root/\"\n",
         "print(PIPELINE_ROOT)\n",
         "\n",
+        "\n",
         "@dsl.pipeline(\n",
         "    name=\"pipeline-trigger-airflow-dag\",\n",
         "    description=\"Trigger Airflow DAG from Vertex AI Pipelines\",\n",
@@ -1140,7 +1161,7 @@
         "    data_processing_task = trigger_airflow_dag(\n",
         "        dag_name=data_processing_task_dag_name,\n",
         "        composer_client_id=COMPOSER_CLIENT_ID,\n",
-        "        composer_webserver_id=COMPOSER_WEB_URI\n",
+        "        composer_webserver_id=COMPOSER_WEB_URI,\n",
         "    )"
       ]
     },
@@ -1171,9 +1192,8 @@
         "    display_name=\"airflow_pipeline\",\n",
         "    template_path=\"pipeline-trigger-airflow-dag.json\",\n",
         "    pipeline_root=PIPELINE_ROOT,\n",
-        "    parameter_values={\n",
-        "    },\n",
-        "    enable_caching=False\n",
+        "    parameter_values={},\n",
+        "    enable_caching=False,\n",
         ")\n",
         "\n",
         "pipeline.run()\n",
@@ -1213,7 +1233,7 @@
       },
       "outputs": [],
       "source": [
-        "COMPOSER_WEB_URI + '/admin/airflow/tree?dag_id=dag_gcs_to_bq_orch'"
+        "COMPOSER_WEB_URI + \"/admin/airflow/tree?dag_id=dag_gcs_to_bq_orch\""
       ]
     },
     {

notebooks/community/ml_ops/stage3/get_started_with_automl_pipeline_components.ipynb

@@ -40,7 +40,7 @@
         "  </td>\n",
         "    <td>\n",
         "        <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage3/get_started_with_automl_pipeline_components.ipynb\">\n",
-        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\\\" alt=\"Colab logo\"> Run in Colab\n",
+        "        <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
         "        </a>\n",
         "  </td>\n",
         "  <td>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with AutoML pipeline components."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower in the given image from the five classes of flowers: daisy, dandelion, rose, sunflower, or tulip."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -100,8 +89,26 @@
         "    - Training a Vertex AI AutoML trained model.\n",
         "    - Test the serving binary with a batch prediction job.\n",
         "    - Deploying a Vertex AI AutoML trained model.\n",
-        "- Execute a Vertex AI pipeline.\n",
+        "- Execute a Vertex AI pipeline.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower in the given image from the five classes of flowers: daisy, dandelion, rose, sunflower, or tulip."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "eef426a35e17"
+      },
+      "source": [
         "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -133,7 +140,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -186,6 +193,8 @@
         "id": "BF1j6f9HApxa"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -196,7 +205,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage3/get_started_with_bq_tfdv_pipeline_components.ipynb

@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with BigQuery and TFDV pipeline components."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:gsod,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -84,7 +73,7 @@
       "source": [
         "### Objective\n",
         "\n",
-        "In this tutorial, you learn how to use build lightweight Python components for BigQuery and Tensorflow Data Validation.\n",
+        "In this tutorial, you learn how to use build lightweight Python components for BigQuery and TensorFlow Data Validation.\n",
         "\n",
         "This tutorial uses the following Google Cloud ML services:\n",
         "\n",
@@ -99,26 +88,31 @@
         "- Execute a Vertex AI pipeline."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:gsod,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "0c997d8d92ce"
       },
       "source": [
-        "### Costs \n",
-        "\n",
-        "\n",
+        "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
-        "* Vertex AI\n",
-        "* Cloud Storage\n",
+        "- Vertex AI\n",
+        "- Cloud Storage\n",
+        "- BigQuery\n",
         "\n",
-        "\n",
-        "Learn about [Vertex AI\n",
-        "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
-        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
-        "Calculator](https://cloud.google.com/products/calculator/)\n",
-        "to generate a cost estimate based on your projected usage."
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and [BigQuery pricing](https://cloud.google.com/bigquery/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
       ]
     },
     {
@@ -143,7 +137,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -193,6 +187,8 @@
         "id": "BF1j6f9HApxa"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -203,7 +199,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,dataflow.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -230,8 +226,6 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
         "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },

notebooks/community/ml_ops/stage3/get_started_with_bqml_pipeline_components.ipynb

@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with BigQuery ML pipeline components."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:penguins,lcn,bq"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the Penguins dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset predicts the species."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -105,26 +94,31 @@
         "- Make a prediction with the deployed Vertex AI model."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:penguins,lcn,bq"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the Penguins dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset predicts the species."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
         "id": "0c997d8d92ce"
       },
       "source": [
-        "### Costs \n",
-        "\n",
-        "\n",
+        "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
-        "* Vertex AI\n",
-        "* Cloud Storage\n",
+        "- Vertex AI\n",
+        "- Cloud Storage\n",
+        "- BigQuery\n",
         "\n",
-        "\n",
-        "Learn about [Vertex AI\n",
-        "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
-        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
-        "Calculator](https://cloud.google.com/products/calculator/)\n",
-        "to generate a cost estimate based on your projected usage."
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and [BigQuery pricing](https://cloud.google.com/bigquery/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
       ]
     },
     {
@@ -149,7 +143,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -201,6 +195,8 @@
         "id": "BF1j6f9HApxa"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -211,7 +207,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,dataflow.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -238,8 +234,6 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
         "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },

notebooks/community/ml_ops/stage3/get_started_with_custom_training_pipeline_components.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 3 : formalization: get started with custom training pipeline components\n",
+        "# E2E ML on GCP: MLOps stage 3 : formalization: get started with Vertex AI custom training pipeline components\n",
         "\n",
         "<table align=\"left\">\n",
         "  <td>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with custom training pipeline components."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:flowers,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -109,6 +98,17 @@
         "- Execute a Vertex AI pipeline."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:flowers,icn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -153,7 +153,7 @@
         "import os\n",
         "\n",
         "# The Vertex AI Workbench Notebook product has specific requirements\n",
-        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\")\n",
+        "IS_WORKBENCH_NOTEBOOK = os.getenv(\"DL_ANACONDA_HOME\") and not os.getenv(\"VIRTUAL_ENV\")\n",
         "IS_USER_MANAGED_WORKBENCH_NOTEBOOK = os.path.exists(\n",
         "    \"/opt/deeplearning/metadata/env_version\"\n",
         ")\n",
@@ -204,6 +204,8 @@
         "id": "BF1j6f9HApxa"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -214,7 +216,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,dataflow.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -241,8 +243,6 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
         "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },

notebooks/community/ml_ops/stage3/get_started_with_dataflow_pipeline_components.ipynb

@@ -67,17 +67,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with Dataflow pipeline components."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:gsod,lrg"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -101,6 +90,39 @@
         "- Execute a Vertex AI pipeline."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:gsod,lrg"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the GSOD dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). The version of the dataset you use only the fields year, month and day to predict the value of mean daily temperature (mean_temp)."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "0c997d8d92ce"
+      },
+      "source": [
+        "### Costs\n",
+        "\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "* Vertex AI\n",
+        "* Cloud Storage\n",
+        "* Dataflow\n",
+        "\n",
+        "Learn about [Vertex AI\n",
+        "pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage\n",
+        "pricing](https://cloud.google.com/storage/pricing), and [Dataflow pricing](https://cloud.google.com/dataflow/pricing)\n",
+        "and use the [Pricing\n",
+        "Calculator](https://cloud.google.com/products/calculator/)\n",
+        "to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -176,6 +198,8 @@
         "id": "BF1j6f9HApxa"
       },
       "source": [
+        "## Before you begin\n",
+        "\n",
         "### Set up your Google Cloud project\n",
         "\n",
         "**The following steps are required, regardless of your notebook environment.**\n",
@@ -186,7 +210,7 @@
         "\n",
         "1. [Enable the Vertex AI API and Dataflow API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,dataflow.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -213,24 +237,7 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
-        "PROJECT_ID = \"\"\n",
-        "\n",
-        "if not os.getenv(\"IS_TESTING\"):\n",
-        "    # Get your Google Cloud project ID from gcloud\n",
-        "    shell_output = !gcloud config list --format 'value(core.project)' 2>/dev/null\n",
-        "    PROJECT_ID = shell_output[0]\n",
-        "    print(\"Project ID: \", PROJECT_ID)"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "37c0a68ff20d"
-      },
-      "source": [
-        "Otherwise, set your project ID here."
+        "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },
     {
@@ -241,8 +248,22 @@
       },
       "outputs": [],
       "source": [
-        "if PROJECT_ID == \"\" or PROJECT_ID is None:\n",
-        "    PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
+        "if PROJECT_ID == \"\" or PROJECT_ID is None or PROJECT_ID == \"[your-project-id]\":\n",
+        "    # Get your GCP project id from gcloud\n",
+        "    shell_output = ! gcloud config list --format 'value(core.project)' 2>/dev/null\n",
+        "    PROJECT_ID = shell_output[0]\n",
+        "    print(\"Project ID:\", PROJECT_ID)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "250cb8c648d5"
+      },
+      "outputs": [],
+      "source": [
+        "! gcloud config set project $PROJECT_ID"
       ]
     },
     {

notebooks/community/ml_ops/stage3/get_started_with_dataproc_serverless_pipeline_components.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 3 : formalization: get started with Datproc Serverless pipeline components\n",
+        "# E2E ML on GCP: MLOps stage 3 : formalization: get started with Dataproc Serverless pipeline components\n",
         "\n",
         "<table align=\"left\">\n",
         "  <td>\n",
@@ -80,12 +80,31 @@
         "- `Google Cloud Pipeline Components`\n",
         "- `Dataproc Serverless`\n",
         "\n",
-        "An example pipeline is provided for each Dataproc Serverless component, which includes:\n",
+        "The steps performed include:\n",
+        "\n",
         "- `DataprocPySparkBatchOp` for running PySpark batch workloads.\n",
         "- `DataprocSparkBatchOp` for running Spark batch workloads.\n",
         "- `DataprocSparkSqlBatchOp` for running Spark SQL batch workloads.\n",
-        "- `DataprocSparkRBatchOp` for running SparkR batch workloads.\n",
+        "- `DataprocSparkRBatchOp` for running SparkR batch workloads."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4ced09c1b4ce"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "No dataset is used in this tutorial. References to an example dataset are for demonstration purposes."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "25697c6fccd3"
+      },
+      "source": [
         "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -96,23 +115,6 @@
         "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Dataproc Serverless pricing](https://cloud.google.com/dataproc-serverless/pricing) and [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "bucket:mbsdk"
-      },
-      "source": [
-        "### Before you begin\n",
-        "\n",
-        "**Before proceeding, you should complete the following pre-requisites:**\n",
-        "\n",
-        "* [Configure your project for Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/configure-project).\n",
-        "\n",
-        "* [Enable the Dataproc API](https://console.cloud.google.com/flows/enableapi?apiid=dataproc.googleleapis.com) in your project.\n",
-        "\n",
-        "* Ensure your project meets the networking requirements detailed in [Dataproc Serverless for Spark network configuration](https://cloud.google.com/dataproc-serverless/docs/concepts/network)."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -197,7 +199,13 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. [Configure your project for Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/configure-project).\n",
+        "\n",
+        "1. [Enable the Dataproc API](https://console.cloud.google.com/flows/enableapi?apiid=dataproc.googleleapis.com) in your project.\n",
+        "\n",
+        "1. Ensure your project meets the networking requirements detailed in [Dataproc Serverless for Spark network configuration](https://cloud.google.com/dataproc-serverless/docs/concepts/network).\n",
+        "\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage3/get_started_with_hpt_pipeline_components.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 3 : formalization: get started with Hyperparameter Tuning pipeline components\n",
+        "# E2E ML on GCP: MLOps stage 3 : formalization: get started with Vertex AI Hyperparameter Tuning pipeline components\n",
         "\n",
         "<table align=\"left\">\n",
         "    <td>\n",
@@ -46,7 +46,7 @@
         "  <td>\n",
         "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/notebooks/community/ml_ops/stage3/get_started_with_hpt_pipeline_components.ipynb\">\n",
         "        <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
-        "      Open in Google Cloud Notebooks\n",
+        "      Open in Vertex AI Workbench\n",
         "    </a>\n",
         "  </td>\n",
         "</table>\n",
@@ -62,18 +62,7 @@
         "## Overview\n",
         "\n",
         "\n",
-        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with Hyperparameter Tuning pipeline components."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:horses_or_humans,icn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Horses or Humans](https://www.tensorflow.org/datasets/catalog/horses_or_humans) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The trained model predicts whether an image is a horse or human being."
+        "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with Vertex AI Hyperparameter Tuning pipeline components."
       ]
     },
     {
@@ -100,8 +89,26 @@
         "    - If the metrics exceed a specified threshold.\n",
         "      - Get the location of the model artifacts for the best tuned model.\n",
         "      - Upload the model artifacts to a `Vertex AI Model` resource.\n",
-        "- Execute a Vertex AI pipeline.\n",
+        "- Execute a Vertex AI pipeline."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:horses_or_humans,icn"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "The dataset used for this tutorial is the [Horses or Humans](https://www.tensorflow.org/datasets/catalog/horses_or_humans) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The trained model predicts whether an image is a horse or human being."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "5e2eba58ad71"
+      },
+      "source": [
         "### Costs \n",
         "\n",
         "This tutorial uses billable components of Google Cloud:\n",
@@ -204,7 +211,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage3/get_started_with_kubeflow_pipelines.ipynb

@@ -87,8 +87,26 @@
         "- Executing a KFP pipeline using Vertex AI Pipelines.\n",
         "- Loading component and pipeline definitions from a source code repository.\n",
         "- Building sequential, parallel, multiple output components.\n",
-        "- Building control flow into pipelines.\n",
+        "- Building control flow into pipelines."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4ced09c1b4ce"
+      },
+      "source": [
+        "### Dataset\n",
         "\n",
+        "No dataset is used in this tutorial. References to an example dataset are for demonstration purposes."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "eef426a35e17"
+      },
+      "source": [
         "### Costs\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
@@ -181,7 +199,7 @@
         "\n",
         "1. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage3/get_started_with_rapid_prototyping_bqml_automl.ipynb

@@ -29,7 +29,7 @@
         "id": "title:generic,gcp"
       },
       "source": [
-        "# E2E ML on GCP: MLOps stage 3 : Get started with rapid prototyping with AutoML and BQML\n",
+        "# E2E ML on GCP: MLOps stage 3 : Get started with rapid prototyping with AutoML and BigQuery ML\n",
         "<table align=\"left\">\n",
         "  <td>\n",
         "    <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage3/get_started_with_rapid_prototyping_bqml_automl.ipynb\">\n",
@@ -65,6 +65,33 @@
         "<img src=\"https://storage.googleapis.com/rafacarv-public-bucket-do-not-delete/abalone/automl_and_bqml.png\" />"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "6c75b63ad57e"
+      },
+      "source": [
+        "### Objective\n",
+        "\n",
+        "In this tutorial, you learn how to use `Vertex AI Predictions` for rapid prototyping a model.\n",
+        "\n",
+        "This tutorial uses the following Google Cloud ML services:\n",
+        "\n",
+        "- `Vertex AI Pipelines`\n",
+        "- `Vertex AI AutoML`\n",
+        "- `Vertex AI BigQuery ML`\n",
+        "- `Google Cloud Pipeline Components`\n",
+        "\n",
+        "The steps performed include:\n",
+        "\n",
+        "- Creating a BigQuery and Vertex AI training dataset.\n",
+        "- Training a BigQuery ML and AutoML model.\n",
+        "- Extracting evaluation metrics from the BigQueryML and AutoML models.\n",
+        "- Selecting the best trained model.\n",
+        "- Deploying the best trained model.\n",
+        "- Testing the deployed model infrastructure."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -153,33 +180,6 @@
         "</body>\n"
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "6c75b63ad57e"
-      },
-      "source": [
-        "### Objective\n",
-        "\n",
-        "In this tutorial, you learn how to use `Vertex AI Predictions` for rapid prototyping a model.\n",
-        "\n",
-        "This tutorial uses the following Google Cloud ML services:\n",
-        "\n",
-        "- `Vertex AI Pipelines`\n",
-        "- `Vertex AI AutoML`\n",
-        "- `Vertex AI BigQuery ML`\n",
-        "- `Google Cloud Pipeline Components`\n",
-        "\n",
-        "The steps performed include:\n",
-        "\n",
-        "- Creating a BigQuery and Vertex AI training dataset.\n",
-        "- Training a BigQuery ML and AutoML model.\n",
-        "- Extracting evaluation metrics from the BigQueryML and AutoML models.\n",
-        "- Selecting the best trained model.\n",
-        "- Deploying the best trained model.\n",
-        "- Testing the deployed model infrastructure."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -187,17 +187,13 @@
       },
       "source": [
         "### Costs\n",
-        "\n",
         "This tutorial uses billable components of Google Cloud:\n",
         "\n",
-        "* Vertex AI\n",
-        "* Cloud Storage\n",
+        "- Vertex AI\n",
+        "- Cloud Storage\n",
+        "- BigQuery\n",
         "\n",
-        "Learn about [Vertex AI\n",
-        "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
-        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
-        "Calculator](https://cloud.google.com/products/calculator/)\n",
-        "to generate a cost estimate based on your projected usage."
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and [BigQuery pricing](https://cloud.google.com/bigquery/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
       ]
     },
     {
@@ -208,7 +204,7 @@
       "source": [
         "### Set up your local development environment\n",
         "\n",
-        "If you are using Colab or Google Cloud Notebooks, your environment already meets all the requirements to run this notebook. You can skip this step.\n",
+        "If you are using Colab or Vertex AI Workbench Notebooks, your environment already meets all the requirements to run this notebook.  \n",
         "\n",
         "Otherwise, make sure your environment meets this notebook's requirements. You need the following:\n",
         "\n",
@@ -300,69 +296,6 @@
         "    app.kernel.do_shutdown(True)"
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "gcp_authenticate"
-      },
-      "source": [
-        "### Authenticate your Google Cloud account\n",
-        "\n",
-        "**If you are using Vertex AI Workbench Notebooks**, your environment is already\n",
-        "authenticated. Skip this step.\n",
-        "\n",
-        "**If you are using Colab**, run the cell below and follow the instructions when prompted to authenticate your account via oAuth.\n",
-        "\n",
-        "\n",
-        "**Otherwise**, follow these steps:\n",
-        "\n",
-        "In the Cloud Console, go to the [Create service account key](https://console.cloud.google.com/apis/credentials/serviceaccountkey) page.\n",
-        "\n",
-        "**Click Create service account**.\n",
-        "\n",
-        "In the **Service account name** field, enter a name, and click **Create**.\n",
-        "\n",
-        "In the **Grant this service account access to project** section, click the Role drop-down list. Type \"Vertex\" into the filter box, and select **Vertex Administrator**. Type \"Storage Object Admin\" into the filter box, and select **Storage Object Admin**.\n",
-        "\n",
-        "Click Create. A JSON file that contains your key downloads to your local environment.\n",
-        "\n",
-        "Enter the path to your service account key as the GOOGLE_APPLICATION_CREDENTIALS variable in the cell below and run the cell."
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "id": "gcp_authenticate"
-      },
-      "outputs": [],
-      "source": [
-        "# If you are running this notebook in Colab, run this cell and follow the\n",
-        "# instructions to authenticate your GCP account. This provides access to your\n",
-        "# Cloud Storage bucket and lets you submit training jobs and prediction\n",
-        "# requests.\n",
-        "\n",
-        "import os\n",
-        "import sys\n",
-        "\n",
-        "# If on Vertex AI Workbench, then don't execute this code\n",
-        "IS_COLAB = \"google.colab\" in sys.modules\n",
-        "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
-        "    \"DL_ANACONDA_HOME\"\n",
-        "):\n",
-        "    if \"google.colab\" in sys.modules:\n",
-        "        IS_COLAB = True\n",
-        "        from google.colab import auth as google_auth\n",
-        "\n",
-        "        google_auth.authenticate_user()\n",
-        "\n",
-        "    # If you are running this notebook locally, replace the string below with the\n",
-        "    # path to your service account key and run this cell to authenticate your GCP\n",
-        "    # account.\n",
-        "    elif not os.getenv(\"IS_TESTING\"):\n",
-        "        %env GOOGLE_APPLICATION_CREDENTIALS ''"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -385,7 +318,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -412,8 +345,6 @@
       },
       "outputs": [],
       "source": [
-        "import os\n",
-        "\n",
         "PROJECT_ID = \"[your-project-id]\"  # @param {type:\"string\"}"
       ]
     },
@@ -500,6 +431,69 @@
         "TIMESTAMP = datetime.now().strftime(\"%Y%m%d%H%M%S\")"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "gcp_authenticate"
+      },
+      "source": [
+        "### Authenticate your Google Cloud account\n",
+        "\n",
+        "**If you are using Vertex AI Workbench Notebooks**, your environment is already\n",
+        "authenticated. Skip this step.\n",
+        "\n",
+        "**If you are using Colab**, run the cell below and follow the instructions when prompted to authenticate your account via oAuth.\n",
+        "\n",
+        "\n",
+        "**Otherwise**, follow these steps:\n",
+        "\n",
+        "In the Cloud Console, go to the [Create service account key](https://console.cloud.google.com/apis/credentials/serviceaccountkey) page.\n",
+        "\n",
+        "**Click Create service account**.\n",
+        "\n",
+        "In the **Service account name** field, enter a name, and click **Create**.\n",
+        "\n",
+        "In the **Grant this service account access to project** section, click the Role drop-down list. Type \"Vertex\" into the filter box, and select **Vertex Administrator**. Type \"Storage Object Admin\" into the filter box, and select **Storage Object Admin**.\n",
+        "\n",
+        "Click Create. A JSON file that contains your key downloads to your local environment.\n",
+        "\n",
+        "Enter the path to your service account key as the GOOGLE_APPLICATION_CREDENTIALS variable in the cell below and run the cell."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "gcp_authenticate"
+      },
+      "outputs": [],
+      "source": [
+        "# If you are running this notebook in Colab, run this cell and follow the\n",
+        "# instructions to authenticate your GCP account. This provides access to your\n",
+        "# Cloud Storage bucket and lets you submit training jobs and prediction\n",
+        "# requests.\n",
+        "\n",
+        "import os\n",
+        "import sys\n",
+        "\n",
+        "# If on Vertex AI Workbench, then don't execute this code\n",
+        "IS_COLAB = \"google.colab\" in sys.modules\n",
+        "if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
+        "    \"DL_ANACONDA_HOME\"\n",
+        "):\n",
+        "    if \"google.colab\" in sys.modules:\n",
+        "        IS_COLAB = True\n",
+        "        from google.colab import auth as google_auth\n",
+        "\n",
+        "        google_auth.authenticate_user()\n",
+        "\n",
+        "    # If you are running this notebook locally, replace the string below with the\n",
+        "    # path to your service account key and run this cell to authenticate your GCP\n",
+        "    # account.\n",
+        "    elif not os.getenv(\"IS_TESTING\"):\n",
+        "        %env GOOGLE_APPLICATION_CREDENTIALS ''"
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {

notebooks/community/ml_ops/stage3/get_started_with_tfx_pipeline.ipynb

@@ -29,7 +29,7 @@
         "id": "ff60de67fa8d"
       },
       "source": [
-        "Notebook is a revised version of an unpublished notebook from Juan Acevedo"
+        "This notebook is a revised version of an unpublished notebook from Juan Acevedo"
       ]
     },
     {
@@ -49,7 +49,7 @@
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samplestree/main/notebooks/community/ml_ops/stage3/get_started_with_tfx_pipeline.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage3/get_started_with_tfx_pipeline.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
@@ -75,17 +75,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization: get started with TFX and Vertex AI Pipelines."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:bank,lbn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [CIFAR10 dataset](https://www.tensorflow.org/datasets/catalog/cifar10) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use is built into TensorFlow. The trained model predicts which type of class an image is from ten classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, or truck."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -112,6 +101,17 @@
         "- Execute the pipeline using `Vertex AI Pipelines`."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:bank,lbn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [CIFAR10 dataset](https://www.tensorflow.org/datasets/catalog/cifar10) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset is built into TensorFlow. The trained model predicts which type of class an image is from ten classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, or truck."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -124,10 +124,12 @@
         "\n",
         "* Vertex AI\n",
         "* Cloud Storage\n",
+        "* Dataflow\n",
         "\n",
         "Learn about [Vertex AI\n",
-        "pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
-        "pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
+        "pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage\n",
+        "pricing](https://cloud.google.com/storage/pricing), and [Dataflow pricing](https://cloud.google.com/dataflow/pricing)\n",
+        "and use the [Pricing\n",
         "Calculator](https://cloud.google.com/products/calculator/)\n",
         "to generate a cost estimate based on your projected usage."
       ]
@@ -225,7 +227,7 @@
         "\n",
         "3. Enable the APIs necessary to execute this notebook -- see cell below.\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -233,6 +235,17 @@
         "**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$`."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "c4ccf556d4ea"
+      },
+      "source": [
+        "### Enable APIs\n",
+        "\n",
+        "You can enable the required APIs using `gcloud`."
+      ]
+    },
     {
       "cell_type": "code",
       "execution_count": null,

notebooks/community/ml_ops/stage3/mlops_formalization.ipynb

@@ -38,7 +38,7 @@
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samplestree/main/notebooks/community/ml_ops/stage3/mlops_formalization.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage3/mlops_formalization.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
@@ -65,17 +65,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 3 : formalization."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:bq,chicago,lbn"
-      },
-      "source": [
-        "### Dataset\n",
-        "\n",
-        "The dataset used for this tutorial is the [Chicago Taxi](https://www.kaggle.com/chicago/chicago-taxi-trips-bq). The version of the dataset you will use in this tutorial is stored in a public BigQuery table. The trained model predicts whether someone would leave a tip for a taxi fare."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -91,7 +80,7 @@
         "- `Vertex AI Pipelines`\n",
         "- `Vertex AI Training`\n",
         "- `Google Cloud Pipeline Components`\n",
-        "- `Vertex AI Dataset, and Model resources\n",
+        "- `Vertex AI Dataset, and Model` resources\n",
         "- `Dataflow`\n",
         "\n",
         "The steps performed include:\n",
@@ -101,7 +90,7 @@
         "    - Dataset schema/statistics for baseline model.\n",
         "- Formalize a data preprocessing pipeline.\n",
         "    - Extract columns/rows from BigQuery table to local BigQuery table.\n",
-        "    - Use Tensorflow Data Validation library to determine statistics, schema, and features.\n",
+        "    - Use TensorFlow Data Validation library to determine statistics, schema, and features.\n",
         "    - Use Dataflow to preprocess the data.\n",
         "    - Create a Vertex AI Dataset.\n",
         "- Formalize a build model architecture pipeline.\n",
@@ -130,7 +119,7 @@
         "   - Training pipeline\n",
         "\n",
         "- The data pipeline should perform the following tasks:\n",
-        "    - Do satistical analysis on the dataset using Tensorflow Data Validation library.\n",
+        "    - Do satistical analysis on the dataset using TensorFlow Data Validation library.\n",
         "    - Split the dataset examples into training, validation and test datasets using `Dataflow` components.\n",
         "    - Preprocess and transform the split datasets into machine learning ready format, i.e., `TFRecord`, using `Dataflow` components.\n",
         "    - Preprocess copies of test dataset for testing serving model using `Dataflow` components.\n",
@@ -156,7 +145,7 @@
         "        - Load and compile the model artifacts.\n",
         "    - Train the model.\n",
         "        - Train the model with corresponding hyperparameters.\n",
-        "        - Track the training with a `Vertex AI Tensorboard` instance.\n",
+        "        - Track the training with a `Vertex AI TensorBoard` instance.\n",
         "        - Store the trained model artifacts on Cloud Storage.\n",
         "    - Evaluate the model.\n",
         "        - Evaluate the model using the test dataset.\n",
@@ -169,6 +158,39 @@
         "    - Deploy the trained `Vertex AI Model` resource to the `Vertex AI Endpoint` resource."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:bq,chicago,lbn"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "The dataset used for this tutorial is the [Chicago Taxi](https://www.kaggle.com/chicago/chicago-taxi-trips-bq). The version of the dataset in this tutorial is stored in a public BigQuery table. The trained model predicts whether someone leaves a tip for a taxi fare."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "0c997d8d92ce"
+      },
+      "source": [
+        "### Costs\n",
+        "\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "* Vertex AI\n",
+        "* Cloud Storage\n",
+        "* Dataflow\n",
+        "\n",
+        "Learn about [Vertex AI\n",
+        "pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage\n",
+        "pricing](https://cloud.google.com/storage/pricing), and [Dataflow pricing](https://cloud.google.com/dataflow/pricing)\n",
+        "and use the [Pricing\n",
+        "Calculator](https://cloud.google.com/products/calculator/)\n",
+        "to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -203,22 +225,22 @@
         "\n",
         "ONCE_ONLY = False\n",
         "if ONCE_ONLY:\n",
-        "    ! pip3 install -U tensorflow==2.5 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-data-validation==1.2 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-transform==1.2 $USER_FLAG\n",
-        "    ! pip3 install -U tensorflow-io==0.18 $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-pipeline-components $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-bigquery $USER_FLAG\n",
-        "    ! pip3 install --upgrade google-cloud-logging $USER_FLAG\n",
-        "    ! pip3 install --upgrade apache-beam[gcp] $USER_FLAG\n",
-        "    ! pip3 install --upgrade pyarrow $USER_FLAG\n",
-        "    ! pip3 install --upgrade cloudml-hypertune $USER_FLAG\n",
-        "    ! pip3 install --upgrade kfp $USER_FLAG\n",
-        "    ! pip3 install --upgrade torchvision $USER_FLAG\n",
-        "    ! pip3 install --upgrade rpy2 $USER_FLAG\n",
-        "    ! pip3 install --upgrade python-tabulate $USER_FLAG\n",
-        "    ! pip3 install -U opencv-python-headless==4.5.2.52 $USER_FLAG"
+        "    ! pip3 install -U tensorflow==2.5 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-data-validation==1.2 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-transform==1.2 $USER_FLAG -q\n",
+        "    ! pip3 install -U tensorflow-io==0.18 $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-aiplatform[tensorboard] $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-pipeline-components $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-bigquery $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade google-cloud-logging $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade apache-beam[gcp] $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade pyarrow $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade cloudml-hypertune $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade kfp $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade torchvision $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade rpy2 $USER_FLAG -q\n",
+        "    ! pip3 install --upgrade python-tabulate $USER_FLAG -q\n",
+        "    ! pip3 install -U opencv-python-headless==4.5.2.52 $USER_FLAG -q"
       ]
     },
     {
@@ -268,7 +290,7 @@
         "\n",
         "1. [Enable the Vertex AI API and Dataflow API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,dataflow.googleapis.com).\n",
         "\n",
-        "1. If you are running this notebook locally, you will need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
+        "1. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk).\n",
         "\n",
         "1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",
@@ -2797,7 +2819,7 @@
         "    - `dataset-id`: The resource ID of the `Dataset` resource to use for training.\n",
         "    - `experiment`: The name of the experiment.\n",
         "    - `run`: The name of the run within this experiment.\n",
-        "    - `tensorboard-logdir`: The logging directory for Vertex AI Tensorboard.\n",
+        "    - `tensorboard-logdir`: The logging directory for Vertex AI TensorBoard.\n",
         "\n",
         "\n",
         "- `get_data()`:\n",

notebooks/community/ml_ops/stage4/README.md

@@ -42,67 +42,10 @@ This stage may be done entirely by MLOps. We recommend:
 
 ### Get Started
 
-[Get started with Google Artifact Registry](get_started_with_google_artifact_registry.ipynb)
-
-```
-The steps performed include:
-
-- Creating a private Docker repository.
-- Tagging a container image, specific to the private Docker repository.
-- Pushing a container image to the private Docker repository.
-- Pulling a container image from the private Docker repository.
-- Deleting a private Docker repository.
-```
-
-Get started with Vertex Model Registry
-
-[Get started with Vertex ML Metadata](get_started_with_vertex_ml_metadata.ipynb)
-
-```
-The steps performed include:
-
-- Create a `Metadatastore` resource.
-- Create (record)/List an `Artifact`, with artifacts and metadata.
-- Create (record)/List an `Execution`.
-- Create (record)/List a `Context`.
-- Add `Artifact` to `Execution` as events.
-- Add `Execution` and `Artifact` into the `Context`
-- Delete `Artifact`, `Execution` and `Context`.
-- Create and run a `Vertex AI Pipeline` ML workflow to train and deploy a scikit-learn model.
-    - Create custom pipeline components that generate artifacts and metadata.
-    - Compare Vertex AI Pipelines runs.
-    - Trace the lineage for pipeline-generated artifacts.
-    - Query your pipeline run metadata.
-```
-
-[Get started with Vertex ML Metadata and AutoML](get_started_with_vertex_ml_metadata_and_automl.ipynb)
-
-```
-The steps performed include:
-
-- Create a `Dataset` resource.
-- Create a corresponding `google.VertexDataset` artifact.
-- Train a model using `AutoML`.
-- Create a corresponding `google.VertexModel` artifact.
-- Create an `Endpoint` resource.
-- Create a corresponding `google.Endpoint` artifact.
-- Deploy the train model to the `Endpoint`.
-- Create an execution and context for the `AutoML` training job and deployment.
-- Add the corresponding artifacts and context to the execution.
-- Add artifact links (event) to the execution.
-- Display the execution graph.
-```
-
-
-Get started with custom model evaluation
-
-Get started with A/B Testing
-
 [Get started with Vertex Explainable AI](get_started_with_vertex_xai.ipynb)
 
 ```
 The steps performed include:
-
 - Train an AutoML tabular model.
     - Do a batch prediction with explanations.
     - Do an online prediction with explanations.
@@ -121,8 +64,82 @@ The steps performed include:
 - Train an custom scikit-learn tabular model.
     - Manually set configuration metadata.
     - Do an online prediction with explanations.
+
 ```
 
+[Get started with Google Artifact Registry](get_started_with_google_artifact_registry.ipynb)
+
+```
+The steps performed include:
+- Creating a private Docker repository.
+- Tagging a container image, specific to the private Docker repository.
+- Pushing a container image to the private Docker repository.
+- Pulling a container image from the private Docker repository.
+- Deleting a private Docker repository.
+```
+
+[Get started with AutoML training and ML Metadata](get_started_with_vertex_ml_metadata_and_automl.ipynb)
+
+```
+The steps performed include:
+- Create a `Dataset` resource.
+- Create a corresponding `google.VertexDataset` artifact.
+- Train a model using `AutoML`.
+- Create a corresponding `google.VertexModel` artifact.
+- Create an `Endpoint` resource.
+- Create a corresponding `google.Endpoint` artifact.
+- Deploy the train model to the `Endpoint`.
+- Create an execution and context for the `AutoML` training job and deployment.
+- Add the corresponding artifacts and context to the execution.
+- Add artifact links (event) to the execution.
+- Display the execution graph.
+```
+
+[Get started with Vertex AI ML Metadata](get_started_with_vertex_ml_metadata.ipynb)
+
+```
+The steps performed include:
+- Create a `Metadatastore` resource.
+- Create (record)/List an `Artifact`, with artifacts and metadata.
+- Create (record)/List an `Execution`.
+- Create (record)/List a `Context`.
+- Add `Artifact` to `Execution` as events.
+- Add `Execution` and `Artifact` into the `Context`
+- Delete `Artifact`, `Execution` and `Context`.
+- Create and run a `Vertex AI Pipeline` ML workflow to train and deploy a scikit-learn model.
+    - Create custom pipeline components that generate artifacts and metadata.
+    - Compare Vertex AI Pipelines runs.
+    - Trace the lineage for pipeline-generated artifacts.
+    - Query your pipeline run metadata.
+```
+
+[Get started with Vertex AI Model Evaluation](get_started_with_model_evaluation.ipynb)
+
+```
+The steps performed include:
+- Evaluate an `AutoML` model.
+    - Train an `AutoML` image classification model.
+    - Retrieve the default evaluation metrics from training.
+    - Do a batch evaluation for a custom evaluation slice.
+- Evaluate a BigQuery ML model.
+    - Train a `BigQuery ML` tabular classification model.
+    - Retrieve the default evaluation metrics from training.
+    - Do a batch evaluation for a custom evaluation slice.
+- Evaluate a custom model.
+    - Do a batch evaluation for a custom evaluation slice.
+    - Add an evaluation to the `Model Registry` for the `Model` resource.
+- Evaluate an `AutoML` model.
+    - Train an `AutoML` image classification model.
+    - Retrieve the default evaluation metrics from training.
+    - Do a batch evaluation for a custom evaluation slice.
+- Evaluate a BigQuery ML model.
+    - Train a `BigQuery ML` tabular classification model.
+    - Retrieve the default evaluation metrics from training.
+    - Do a batch evaluation for a custom evaluation slice.
+- Evaluate a custom model.
+    - Do a batch evaluation for a custom evaluation slice.
+    - Add an evaluation to the `Model Registry` for the `Model` resource.
+```
 ### E2E Stage Example
 
 Stage 4: Evaluation

notebooks/community/ml_ops/stage4/get_started_with_google_artifact_registry.ipynb

@@ -88,6 +88,33 @@
         "- Deleting a private Docker repository."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4ced09c1b4ce"
+      },
+      "source": [
+        "### Dataset\n",
+        "\n",
+        "No dataset is used in this tutorial. References to an example dataset are for demonstration purposes."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "35bee437737d"
+      },
+      "source": [
+        "### Costs\n",
+        "\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "* Vertex AI\n",
+        "* Cloud Storage\n",
+        "\n",
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage pricing](https://cloud.google.com/storage/pricing), and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -174,7 +201,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",

notebooks/community/ml_ops/stage4/get_started_with_model_evaluation.ipynb

@@ -40,7 +40,7 @@
         "    </a>\n",
         "  </td>\n",
         "  <td>\n",
-        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samplestree/main/notebooks/community/ml_ops/stage4/get_started_with_model_evaluation.ipynb\">\n",
+        "    <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/ml_ops/stage4/get_started_with_model_evaluation.ipynb\">\n",
         "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
         "      View on GitHub\n",
         "    </a>\n",
@@ -66,35 +66,6 @@
         "This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 4 : evaluation: get started with Vertex AI Model Evaluation."
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "id": "dataset:bank,lbn"
-      },
-      "source": [
-        "### Datasets\n",
-        "\n",
-        "**AutoML image model**\n",
-        "\n",
-        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip.\n",
-        "\n",
-        "**BigQuery ML tabular model**\n",
-        "\n",
-        "The dataset used for this tutorial is the Penguins dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). This version of the dataset is used to predict the species of penguins from the available features like culmen-length, flipper-depth etc.\n",
-        "\n",
-        "**Custom model**\n",
-        "\n",
-        "This tutorial uses a pre-trained image classification model from TensorFlow Hub, which is trained on ImageNet dataset.\n",
-        "\n",
-        "Learn more about [ResNet V2 pretained model](https://tfhub.dev/google/imagenet/resnet_v2_101/classification/5). \n",
-        "\n",
-        "\n",
-        "**Pipeline**\n",
-        "BLAH\n",
-        "\n",
-        "The dataset used for this tutorial is the [Bank Marketing](https://pantheon.corp.google.com/storage/browser/_details/cloud-ml-tables-data/bank-marketing.csv) . This dataset does not require any feature engineering. The version of the dataset you will use in this tutorial is stored in a public Cloud Storage bucket."
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -145,6 +116,50 @@
         "    - Add an evaluation to the `Model Registry` for the `Model` resource."
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dataset:bank,lbn"
+      },
+      "source": [
+        "### Datasets\n",
+        "\n",
+        "**AutoML image model**\n",
+        "\n",
+        "The dataset used for this tutorial is the [Flowers dataset](https://www.tensorflow.org/datasets/catalog/tf_flowers) from [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview). The version of the dataset in this tutorial is stored in a public Cloud Storage bucket. The trained model predicts the type of flower an image is from a class of five flowers: daisy, dandelion, rose, sunflower, or tulip.\n",
+        "\n",
+        "**BigQuery ML tabular model**\n",
+        "\n",
+        "The dataset used for this tutorial is the Penguins dataset from [BigQuery public datasets](https://cloud.google.com/bigquery/public-data). This version of the dataset is used to predict the species of penguins from the available features like culmen-length, flipper-depth etc.\n",
+        "\n",
+        "**Custom model**\n",
+        "\n",
+        "This tutorial uses a pre-trained image classification model from TensorFlow Hub, which is trained on ImageNet dataset.\n",
+        "\n",
+        "Learn more about [ResNet V2 pretained model](https://tfhub.dev/google/imagenet/resnet_v2_101/classification/5). \n",
+        "\n",
+        "\n",
+        "**Pipeline**\n",
+        "\n",
+        "The dataset used for this tutorial is the [Bank Marketing](https://pantheon.corp.google.com/storage/browser/_details/cloud-ml-tables-data/bank-marketing.csv) . This dataset does not require any feature engineering. The version of the dataset in this tutorial is stored in a public Cloud Storage bucket."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "0c997d8d92ce"
+      },
+      "source": [
+        "### Costs\n",
+        "This tutorial uses billable components of Google Cloud:\n",
+        "\n",
+        "- Vertex AI\n",
+        "- Cloud Storage\n",
+        "- BigQuery\n",
+        "\n",
+        "Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing), [Cloud Storage pricing](https://cloud.google.com/storage/pricing) and [BigQuery pricing](https://cloud.google.com/bigquery/pricing) and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
+      ]
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -236,7 +251,7 @@
         "\n",
         "3. [Enable the following APIs: Vertex AI APIs, Compute Engine APIs, and Cloud Storage.](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,storage-component.googleapis.com)\n",
         "\n",
-        "4. If you are running this notebook locally, you will need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
+        "4. If you are running this notebook locally, you need to install the [Cloud SDK]((https://cloud.google.com/sdk)).\n",
         "\n",
         "5. Enter your project ID in the cell below. Then run the  cell to make sure the\n",
         "Cloud SDK uses the right project for all the commands in this notebook.\n",