google-cloud-pipeline-components 2.15.0__py3-none-any.whl → 2.16.1__py3-none-any.whl

google_cloud_pipeline_components/preview/custom_job/utils.py

@@ -54,7 +54,7 @@ def create_custom_training_job_from_component(
     display_name: str = '',
     replica_count: int = 1,
     machine_type: str = 'n1-standard-4',
-    accelerator_type: str = '',
+    accelerator_type: str = 'ACCELERATOR_TYPE_UNSPECIFIED',
     accelerator_count: int = 1,
     boot_disk_type: str = 'pd-ssd',
     boot_disk_size_gb: int = 100,
@@ -83,9 +83,9 @@ def create_custom_training_job_from_component(
       replica_count: The count of instances in the cluster. One replica always counts towards the master in worker_pool_spec[0] and the remaining replicas will be allocated in worker_pool_spec[1]. See [more information.](https://cloud.google.com/vertex-ai/docs/training/distributed-training#configure_a_distributed_training_job)
       machine_type: The type of the machine to run the CustomJob. The default value is "n1-standard-4". See [more information](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types).
       accelerator_type: The type of accelerator(s) that may be attached to the machine per `accelerator_count`. See [more information](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/MachineSpec#acceleratortype).
-      accelerator_count: The number of accelerators to attach to the machine. Defaults to 1 if `accelerator_type` is set.
-      boot_disk_type: Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive). boot_disk_type is set as a static value and cannot be changed as a pipeline parameter.
-      boot_disk_size_gb: Size in GB of the boot disk (default is 100GB). `boot_disk_size_gb` is set as a static value and cannot be changed as a pipeline parameter.
+      accelerator_count: The number of accelerators to attach to the machine. Defaults to 1 if `accelerator_type` is set statically.
+      boot_disk_type: Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
+      boot_disk_size_gb: Size in GB of the boot disk (default is 100GB).
       timeout: The maximum job running time. The default is 7 days. A duration in seconds with up to nine fractional digits, terminated by 's', for example: "3.5s".
       restart_job_on_worker_restart: Restarts the entire CustomJob if a worker gets restarted. This feature can be used by distributed training jobs that are not resilient to workers leaving and joining a job.
       service_account: Sets the default service account for workload run-as account. The [service account](https://cloud.google.com/vertex-ai/docs/pipelines/configure-project#service-account) running the pipeline submitting jobs must have act-as permission on this run-as account. If unspecified, the Vertex AI Custom Code [Service Agent](https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) for the CustomJob's project.
@@ -94,11 +94,11 @@ def create_custom_training_job_from_component(
       tensorboard: The name of a Vertex AI TensorBoard resource to which this CustomJob will upload TensorBoard logs.
       enable_web_access: Whether you want Vertex AI to enable [interactive shell access](https://cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) to training containers. If `True`, you can access interactive shells at the URIs given by [CustomJob.web_access_uris][].
       reserved_ip_ranges: A list of names for the reserved IP ranges under the VPC network that can be used for this job. If set, we will deploy the job within the provided IP ranges. Otherwise, the job will be deployed to any IP ranges under the provided VPC network.
-      nfs_mounts: A list of [NfsMount](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/CustomJobSpec#NfsMount) resource specs in Json dict format. For more details about mounting NFS for CustomJob, see [Mount an NFS share for custom training](https://cloud.google.com/vertex-ai/docs/training/train-nfs-share).
+      nfs_mounts: A list of [NfsMount](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/CustomJobSpec#NfsMount) resource specs in Json dict format. For more details about mounting NFS for CustomJob, see [Mount an NFS share for custom training](https://cloud.google.com/vertex-ai/docs/training/train-nfs-share). `nfs_mounts` is set as a static value and cannot be changed as a pipeline parameter.
       base_output_directory: The Cloud Storage location to store the output of this CustomJob or HyperparameterTuningJob. See [more information](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/GcsDestination).
       labels: The labels with user-defined metadata to organize the CustomJob. See [more information](https://goo.gl/xmQnxf).
       persistent_resource_id: The ID of the PersistentResource in the same Project and Location which to run. The default value is a placeholder that will be resolved to the PipelineJob [RuntimeConfig](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.pipelineJobs#PipelineJob.RuntimeConfig)'s persistent resource id at runtime. However, if the PipelineJob doesn't set Persistent Resource as the job level runtime, the placedholder will be resolved to an empty string and the custom job will be run on demand. If the value is set explicitly, the custom job will runs in the specified persistent resource, in this case, please note the network and CMEK configs on the job should be consistent with those on the PersistentResource, otherwise, the job will be rejected. (This is a Preview feature not yet recommended for production workloads.)
-      env: Environment variables to be passed to the container. Takes the form `[{'name': '...', 'value': '...'}]`. Maximum limit is 100.
+      env: Environment variables to be passed to the container. Takes the form `[{'name': '...', 'value': '...'}]`. Maximum limit is 100. `env` is set as a static value and cannot be changed as a pipeline parameter.
 
   Returns:
       A KFP component with CustomJob specification applied.
@@ -148,7 +148,11 @@ def create_custom_training_job_from_component(
   )[0]['container']
 
   worker_pool_spec = {
-      'machine_spec': {'machine_type': machine_type},
+      'machine_spec': {
+          'machine_type': "{{$.inputs.parameters['machine_type']}}",
+          'accelerator_type': "{{$.inputs.parameters['accelerator_type']}}",
+          'accelerator_count': "{{$.inputs.parameters['accelerator_count']}}",
+      },
       'replica_count': 1,
       'container_spec': {
           'image_uri': user_component_container['image'],
@@ -160,15 +164,11 @@ def create_custom_training_job_from_component(
           ),
           'env': env or [],
       },
+      'disk_spec': {
+          'boot_disk_type': "{{$.inputs.parameters['boot_disk_type']}}",
+          'boot_disk_size_gb': "{{$.inputs.parameters['boot_disk_size_gb']}}",
+      },
   }
-  if accelerator_type:
-    worker_pool_spec['machine_spec']['accelerator_type'] = accelerator_type
-    worker_pool_spec['machine_spec']['accelerator_count'] = accelerator_count
-  if boot_disk_type:
-    worker_pool_spec['disk_spec'] = {
-        'boot_disk_type': boot_disk_type,
-        'boot_disk_size_gb': boot_disk_size_gb,
-    }
   if nfs_mounts:
     worker_pool_spec['nfs_mounts'] = nfs_mounts
 
@@ -210,6 +210,54 @@ def create_custom_training_job_from_component(
         'defaultValue'
     ] = default_value
 
+  # add workerPoolSpec parameters into the customjob component
+  cj_component_spec['inputDefinitions']['parameters']['machine_type'] = {
+      'parameterType': 'STRING',
+      'defaultValue': machine_type,
+      'isOptional': True,
+  }
+  cj_component_spec['inputDefinitions']['parameters']['accelerator_type'] = {
+      'parameterType': 'STRING',
+      'defaultValue': accelerator_type,
+      'isOptional': True,
+  }
+  cj_component_spec['inputDefinitions']['parameters']['accelerator_count'] = {
+      'parameterType': 'NUMBER_INTEGER',
+      'defaultValue': (
+          accelerator_count
+          if accelerator_type != 'ACCELERATOR_TYPE_UNSPECIFIED'
+          else 0
+      ),
+      'isOptional': True,
+  }
+  cj_component_spec['inputDefinitions']['parameters']['boot_disk_type'] = {
+      'parameterType': 'STRING',
+      'defaultValue': boot_disk_type,
+      'isOptional': True,
+  }
+  cj_component_spec['inputDefinitions']['parameters']['boot_disk_size_gb'] = {
+      'parameterType': 'NUMBER_INTEGER',
+      'defaultValue': boot_disk_size_gb,
+      'isOptional': True,
+  }
+
+  # check if user component has any input parameters that already exist in the
+  # custom job component
+  for param_name in user_component_spec.get('inputDefinitions', {}).get(
+      'parameters', {}
+  ):
+    if param_name in cj_component_spec['inputDefinitions']['parameters']:
+      raise ValueError(
+          f'Input parameter {param_name} already exists in the CustomJob component.'  # pylint: disable=line-too-long
+      )
+  for param_name in user_component_spec.get('outputDefinitions', {}).get(
+      'parameters', {}
+  ):
+    if param_name in cj_component_spec['outputDefinitions']['parameters']:
+      raise ValueError(
+          f'Output parameter {param_name} already exists in the CustomJob component.'  # pylint: disable=line-too-long
+      )
+
   # merge parameters from user component into the customjob component
   cj_component_spec['inputDefinitions']['parameters'].update(
       user_component_spec.get('inputDefinitions', {}).get('parameters', {})
@@ -217,6 +265,7 @@ def create_custom_training_job_from_component(
   cj_component_spec['outputDefinitions']['parameters'].update(
       user_component_spec.get('outputDefinitions', {}).get('parameters', {})
   )
+
   # use artifacts from user component
   ## assign artifacts, not update, since customjob has no artifact outputs
   cj_component_spec['inputDefinitions']['artifacts'] = user_component_spec.get(

google_cloud_pipeline_components/preview/starry_net/component.py

@@ -57,6 +57,8 @@ def starry_net(  # pylint: disable=dangerous-default-value
     dataprep_target_column: str = '',
     dataprep_static_covariate_columns: List[str] = [],
     dataprep_previous_run_dir: str = '',
+    dataprep_nan_threshold: float = 0.2,
+    dataprep_zero_threshold: float = 0.2,
     trainer_machine_type: str = 'n1-standard-4',
     trainer_accelerator_type: str = 'NVIDIA_TESLA_V100',
     trainer_num_epochs: int = 50,
@@ -84,7 +86,16 @@ def starry_net(  # pylint: disable=dangerous-default-value
     project: str = _placeholders.PROJECT_ID_PLACEHOLDER,
 ):
   # fmt: off
-  """Trains a STARRY-Net model.
+  """Starry Net is a state-of-the-art forecaster used internally by Google.
+
+  Starry Net is a glass-box neural network inspired by statistical time series
+  models, capable of cleaning step changes and spikes, modeling seasonality and
+  events, forecasting trend, and providing both point and prediction interval
+  forecasts in a single, lightweight model. Starry Net stands out among neural
+  network based forecasting models by providing the explainability,
+  interpretability and tunability of traditional statistical forecasters.
+  For example, it features time series feature decomposition and damped local
+  linear exponential smoothing model as the trend structure.
 
   Args:
     tensorboard_instance_id: The tensorboard instance ID. This must be in same
@@ -149,6 +160,13 @@ def starry_net(  # pylint: disable=dangerous-default-value
     dataprep_previous_run_dir: The dataprep dir from a previous run. Use this
       to save time if you've already created TFRecords from your BigQuery
       dataset with the same dataprep parameters as this run.
+    dataprep_nan_threshold: Series having more nan / missing values than
+      nan_threshold (inclusive) in percentage for either backtest or forecast
+      will not be sampled in the training set (including missing due to
+      train_start and train_end). All existing nans are replaced by zeros.
+    dataprep_zero_threshold: Series having more 0.0 values than zero_threshold
+      (inclusive) in percentage for either backtest or forecast will not be
+      sampled in the training set.
     trainer_machine_type: The machine type for training. Must be compatible with
       trainer_accelerator_type.
     trainer_accelerator_type: The accelerator type for training.
@@ -221,39 +239,6 @@ def starry_net(  # pylint: disable=dangerous-default-value
       model_blocks=trainer_model_blocks,
       static_covariates=dataprep_static_covariate_columns,
   )
-  test_set_task = DataprepOp(
-      backcast_length=dataprep_backcast_length,
-      forecast_length=dataprep_forecast_length,
-      train_end_date=dataprep_train_end_date,
-      n_val_windows=dataprep_n_val_windows,
-      n_test_windows=dataprep_n_test_windows,
-      test_set_stride=dataprep_test_set_stride,
-      model_blocks=create_dataprep_args_task.outputs['model_blocks'],
-      bigquery_source=dataprep_bigquery_data_path,
-      ts_identifier_columns=create_dataprep_args_task.outputs[
-          'ts_identifier_columns'],
-      time_column=dataprep_time_column,
-      static_covariate_columns=create_dataprep_args_task.outputs[
-          'static_covariate_columns'],
-      target_column=dataprep_target_column,
-      machine_type=dataflow_machine_type,
-      docker_region=create_dataprep_args_task.outputs['docker_region'],
-      location=location,
-      project=project,
-      job_id=job_id,
-      job_name_prefix='test-set',
-      num_workers=dataflow_starting_replica_count,
-      max_num_workers=dataflow_max_replica_count,
-      disk_size_gb=dataflow_disk_size_gb,
-      test_set_only=True,
-      bigquery_output=dataprep_test_set_bigquery_dataset,
-      gcs_source=dataprep_csv_data_path,
-      gcs_static_covariate_source=dataprep_csv_static_covariates_path,
-      encryption_spec_key_name=encryption_spec_key_name
-  )
-  test_set_task.set_display_name('create-test-set')
-  set_test_set_task = SetTestSetOp(
-      dataprep_dir=test_set_task.outputs['dataprep_dir'])
   with dsl.If(create_dataprep_args_task.outputs['create_tf_records'] == True,  # pylint: disable=singleton-comparison
               'create-tf-records'):
     create_tf_records_task = DataprepOp(
@@ -270,6 +255,7 @@ def starry_net(  # pylint: disable=dangerous-default-value
         time_column=dataprep_time_column,
         static_covariate_columns=create_dataprep_args_task.outputs[
             'static_covariate_columns'],
+        static_covariates_vocab_path='',
         target_column=dataprep_target_column,
         machine_type=dataflow_machine_type,
         docker_region=create_dataprep_args_task.outputs['docker_region'],
@@ -282,6 +268,8 @@ def starry_net(  # pylint: disable=dangerous-default-value
         disk_size_gb=dataflow_disk_size_gb,
         test_set_only=False,
         bigquery_output=dataprep_test_set_bigquery_dataset,
+        nan_threshold=dataprep_nan_threshold,
+        zero_threshold=dataprep_zero_threshold,
         gcs_source=dataprep_csv_data_path,
         gcs_static_covariate_source=dataprep_csv_static_covariates_path,
         encryption_spec_key_name=encryption_spec_key_name
@@ -303,6 +291,42 @@ def starry_net(  # pylint: disable=dangerous-default-value
           'static_covariates_vocab_path'],
       set_tfrecord_args_this_run_task.outputs['static_covariates_vocab_path']
   )
+  test_set_task = DataprepOp(
+      backcast_length=dataprep_backcast_length,
+      forecast_length=dataprep_forecast_length,
+      train_end_date=dataprep_train_end_date,
+      n_val_windows=dataprep_n_val_windows,
+      n_test_windows=dataprep_n_test_windows,
+      test_set_stride=dataprep_test_set_stride,
+      model_blocks=create_dataprep_args_task.outputs['model_blocks'],
+      bigquery_source=dataprep_bigquery_data_path,
+      ts_identifier_columns=create_dataprep_args_task.outputs[
+          'ts_identifier_columns'],
+      time_column=dataprep_time_column,
+      static_covariate_columns=create_dataprep_args_task.outputs[
+          'static_covariate_columns'],
+      static_covariates_vocab_path=static_covariates_vocab_path,
+      target_column=dataprep_target_column,
+      machine_type=dataflow_machine_type,
+      docker_region=create_dataprep_args_task.outputs['docker_region'],
+      location=location,
+      project=project,
+      job_id=job_id,
+      job_name_prefix='test-set',
+      num_workers=dataflow_starting_replica_count,
+      max_num_workers=dataflow_max_replica_count,
+      disk_size_gb=dataflow_disk_size_gb,
+      test_set_only=True,
+      bigquery_output=dataprep_test_set_bigquery_dataset,
+      nan_threshold=dataprep_nan_threshold,
+      zero_threshold=dataprep_zero_threshold,
+      gcs_source=dataprep_csv_data_path,
+      gcs_static_covariate_source=dataprep_csv_static_covariates_path,
+      encryption_spec_key_name=encryption_spec_key_name
+  )
+  test_set_task.set_display_name('create-test-set')
+  set_test_set_task = SetTestSetOp(
+      dataprep_dir=test_set_task.outputs['dataprep_dir'])
   train_tf_record_patterns = dsl.OneOf(
       set_tfrecord_args_previous_run_task.outputs['train_tf_record_patterns'],
       set_tfrecord_args_this_run_task.outputs['train_tf_record_patterns']
@@ -330,6 +354,8 @@ def starry_net(  # pylint: disable=dangerous-default-value
       n_val_windows=dataprep_n_val_windows,
       n_test_windows=dataprep_n_test_windows,
       test_set_stride=dataprep_test_set_stride,
+      nan_threshold=dataprep_nan_threshold,
+      zero_threshold=dataprep_zero_threshold,
       cleaning_activation_regularizer_coeff=trainer_cleaning_activation_regularizer_coeff,
       change_point_activation_regularizer_coeff=trainer_change_point_activation_regularizer_coeff,
       change_point_output_regularizer_coeff=trainer_change_point_output_regularizer_coeff,

google_cloud_pipeline_components/proto/template_metadata_pb2.py

@@ -1,11 +1,13 @@
 # -*- coding: utf-8 -*-
 # Generated by the protocol buffer compiler.  DO NOT EDIT!
-# Protobuf Python Version: 0.20240110.0
+# NO CHECKED-IN PROTOBUF GENCODE
+# Protobuf Python Version: 0.20240806.0
 """Generated protocol buffer code."""
 from google.protobuf import descriptor as _descriptor
 from google.protobuf import descriptor_pool as _descriptor_pool
 from google.protobuf import symbol_database as _symbol_database
 from google.protobuf.internal import builder as _builder
+
 # @@protoc_insertion_point(imports)
 
 _sym_db = _symbol_database.Default()
@@ -71,14 +73,16 @@ DESCRIPTOR = _descriptor_pool.Default().AddSerializedFile(
     b' \x03(\x0b\x32\x36.template_metadata.GoogleCloudServiceAccountValidation\x12O\n\x11quota_validations\x18\x02'
     b' \x03(\x0b\x32\x34.template_metadata.GoogleCloudProjectQuotaValidation\x12N\n\x0f\x61pi_validations\x18\x03'
     b' \x03(\x0b\x32\x35.template_metadata.GoogleCloudApiEnablementValidation\x12H\n\x0fgcs_validations\x18\x04'
-    b' \x03(\x0b\x32/.template_metadata.GoogleCloudStorageValidation"\x80\x01\n\x1cGoogleCloudStorageValidation\x12\x0f\n\x07gcs_uri\x18\x01'
+    b' \x03(\x0b\x32/.template_metadata.GoogleCloudStorageValidation"\x92\x01\n\x1cGoogleCloudStorageValidation\x12\x0f\n\x07gcs_uri\x18\x01'
     b' \x01(\t\x12\x10\n\x08is_input\x18\x02'
     b' \x01(\x08\x12\x1f\n\x17\x64\x65\x66\x61ult_service_account\x18\x03'
     b' \x01(\t\x12\x1c\n\x14override_placeholder\x18\x04'
-    b' \x01(\t"p\n!GoogleCloudProjectQuotaValidation\x12\x13\n\x0bmetric_name\x18\x01'
+    b' \x01(\t\x12\x10\n\x08gcs_uris\x18\x05'
+    b' \x03(\t"\x80\x01\n!GoogleCloudProjectQuotaValidation\x12\x13\n\x0bmetric_name\x18\x01'
     b' \x01(\t\x12\x15\n\x0bint64_value\x18\x02'
     b' \x01(\x03H\x00\x12\x16\n\x0c\x64ouble_value\x18\x03'
-    b' \x01(\x01H\x00\x42\x07\n\x05value"\x8d\x01\n#GoogleCloudServiceAccountValidation\x12\x1f\n\x17\x64\x65\x66\x61ult_principal_email\x18\x01'
+    b' \x01(\x01H\x00\x12\x0e\n\x06region\x18\x04'
+    b' \x01(\tB\x07\n\x05value"\x8d\x01\n#GoogleCloudServiceAccountValidation\x12\x1f\n\x17\x64\x65\x66\x61ult_principal_email\x18\x01'
     b' \x01(\t\x12\x1c\n\x14override_placeholder\x18\x02'
     b' \x01(\t\x12\x13\n\x0bpermissions\x18\x03'
     b' \x03(\t\x12\x12\n\nrole_names\x18\x04'
@@ -96,12 +100,12 @@ _builder.BuildTopDescriptorsAndMessages(
 if not _descriptor._USE_C_DESCRIPTORS:
   _globals['DESCRIPTOR']._loaded_options = None
   _globals['DESCRIPTOR']._serialized_options = b'P\001'
-  _globals['_SIZE']._serialized_start = 3092
-  _globals['_SIZE']._serialized_end = 3163
-  _globals['_CONTENTTYPE']._serialized_start = 3166
-  _globals['_CONTENTTYPE']._serialized_end = 3296
-  _globals['_URITYPE']._serialized_start = 3298
-  _globals['_URITYPE']._serialized_end = 3395
+  _globals['_SIZE']._serialized_start = 3127
+  _globals['_SIZE']._serialized_end = 3198
+  _globals['_CONTENTTYPE']._serialized_start = 3201
+  _globals['_CONTENTTYPE']._serialized_end = 3331
+  _globals['_URITYPE']._serialized_start = 3333
+  _globals['_URITYPE']._serialized_end = 3430
   _globals['_TEMPLATEMETADATA']._serialized_start = 164
   _globals['_TEMPLATEMETADATA']._serialized_end = 301
   _globals['_IOMETADATA']._serialized_start = 303
@@ -143,11 +147,11 @@ if not _descriptor._USE_C_DESCRIPTORS:
   _globals['_VALIDATIONITEMS']._serialized_start = 2308
   _globals['_VALIDATIONITEMS']._serialized_end = 2640
   _globals['_GOOGLECLOUDSTORAGEVALIDATION']._serialized_start = 2643
-  _globals['_GOOGLECLOUDSTORAGEVALIDATION']._serialized_end = 2771
-  _globals['_GOOGLECLOUDPROJECTQUOTAVALIDATION']._serialized_start = 2773
-  _globals['_GOOGLECLOUDPROJECTQUOTAVALIDATION']._serialized_end = 2885
-  _globals['_GOOGLECLOUDSERVICEACCOUNTVALIDATION']._serialized_start = 2888
-  _globals['_GOOGLECLOUDSERVICEACCOUNTVALIDATION']._serialized_end = 3029
-  _globals['_GOOGLECLOUDAPIENABLEMENTVALIDATION']._serialized_start = 3031
-  _globals['_GOOGLECLOUDAPIENABLEMENTVALIDATION']._serialized_end = 3090
+  _globals['_GOOGLECLOUDSTORAGEVALIDATION']._serialized_end = 2789
+  _globals['_GOOGLECLOUDPROJECTQUOTAVALIDATION']._serialized_start = 2792
+  _globals['_GOOGLECLOUDPROJECTQUOTAVALIDATION']._serialized_end = 2920
+  _globals['_GOOGLECLOUDSERVICEACCOUNTVALIDATION']._serialized_start = 2923
+  _globals['_GOOGLECLOUDSERVICEACCOUNTVALIDATION']._serialized_end = 3064
+  _globals['_GOOGLECLOUDAPIENABLEMENTVALIDATION']._serialized_start = 3066
+  _globals['_GOOGLECLOUDAPIENABLEMENTVALIDATION']._serialized_end = 3125
 # @@protoc_insertion_point(module_scope)

google_cloud_pipeline_components/v1/automl/forecasting/bqml_arima_predict_pipeline.yaml

@@ -658,7 +658,7 @@ deploymentSpec:
           \ = client.create_dataset(dataset=dataset, exists_ok=exists_ok)\n  return\
           \ collections.namedtuple('Outputs', ['project_id', 'dataset_id'])(\n   \
           \   ref.project, ref.dataset_id)\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-bigquery-create-dataset-2:
       container:
         args:
@@ -693,7 +693,7 @@ deploymentSpec:
           \ = client.create_dataset(dataset=dataset, exists_ok=exists_ok)\n  return\
           \ collections.namedtuple('Outputs', ['project_id', 'dataset_id'])(\n   \
           \   ref.project, ref.dataset_id)\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-bigquery-delete-dataset-with-prefix:
       container:
         args:
@@ -727,7 +727,7 @@ deploymentSpec:
           \    if dataset.dataset_id.startswith(dataset_prefix):\n      client.delete_dataset(\n\
           \          dataset=dataset.dataset_id,\n          delete_contents=delete_contents)\n\
           \n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-bigquery-query-job:
       container:
         args:
@@ -788,7 +788,7 @@ deploymentSpec:
           \        'datasetId': dataset_id,\n        'tableId': table_id,\n    }\n\
           \  if write_disposition:\n    config['write_disposition'] = write_disposition\n\
           \  return config\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-get-first-valid:
       container:
         args:
@@ -812,7 +812,7 @@ deploymentSpec:
           \  import json\n  # pylint: enable=g-import-not-at-top,import-outside-toplevel,redefined-outer-name,reimported\n\
           \n  for value in json.loads(values):\n    if value:\n      return value\n\
           \  raise ValueError('No valid values.')\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-get-model-metadata:
       container:
         args:
@@ -851,7 +851,7 @@ deploymentSpec:
           \          'forecast_horizon',\n      ],\n  )(\n      options.time_series_timestamp_column,\n\
           \      options.time_series_id_column,\n      options.time_series_data_column,\n\
           \      options.horizon,\n  )\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-get-table-location:
       container:
         args:
@@ -887,7 +887,7 @@ deploymentSpec:
           \  if table.startswith('bq://'):\n    table = table[len('bq://'):]\n  elif\
           \ table.startswith('bigquery://'):\n    table = table[len('bigquery://'):]\n\
           \  return client.get_table(table).location\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-load-table-from-uri:
       container:
         args:
@@ -928,7 +928,7 @@ deploymentSpec:
           \ source_format=source_format)\n  client.load_table_from_uri(\n      source_uris=csv_list,\n\
           \      destination=destination,\n      project=project,\n      location=location,\n\
           \      job_config=job_config).result()\n  return destination\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-maybe-replace-with-default:
       container:
         args:
@@ -950,7 +950,7 @@ deploymentSpec:
           \ *\n\ndef maybe_replace_with_default(value: str, default: str = '') ->\
           \ str:\n  \"\"\"Replaces string with another value if it is a dash.\"\"\"\
           \n  return default if not value else value\n\n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
     exec-validate-inputs:
       container:
         args:
@@ -1046,7 +1046,7 @@ deploymentSpec:
           \   raise ValueError(\n          'Granularity unit should be one of the\
           \ following: '\n          f'{valid_data_granularity_units}, got: {data_granularity_unit}.')\n\
           \n"
-        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240419_0625
+        image: us-docker.pkg.dev/vertex-ai/automl-tabular/kfp-v2-base:20240808_0625
 pipelineInfo:
   description: Forecasts using a BQML ARIMA_PLUS model.
   name: automl-tabular-bqml-arima-prediction