#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you 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
#
# http://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.
"""This module contains a Google Cloud Vertex AI hook."""
from __future__ import annotations
import asyncio
from collections.abc import Sequence
from typing import TYPE_CHECKING, Any
from google.api_core.client_options import ClientOptions
from google.api_core.gapic_v1.method import DEFAULT, _MethodDefault
from google.cloud.aiplatform import (
CustomContainerTrainingJob,
CustomPythonPackageTrainingJob,
CustomTrainingJob,
datasets,
models,
)
from google.cloud.aiplatform_v1 import (
JobServiceAsyncClient,
JobServiceClient,
JobState,
PipelineServiceAsyncClient,
PipelineServiceClient,
PipelineState,
types,
)
from airflow.exceptions import AirflowException, AirflowProviderDeprecationWarning
from airflow.providers.google.common.consts import CLIENT_INFO
from airflow.providers.google.common.deprecated import deprecated
from airflow.providers.google.common.hooks.base_google import GoogleBaseAsyncHook, GoogleBaseHook
if TYPE_CHECKING:
from google.api_core.operation import Operation
from google.api_core.retry import AsyncRetry, Retry
from google.auth.credentials import Credentials
from google.cloud.aiplatform_v1.services.job_service.pagers import ListCustomJobsPager
from google.cloud.aiplatform_v1.services.pipeline_service.pagers import (
ListPipelineJobsPager,
ListTrainingPipelinesPager,
)
from google.cloud.aiplatform_v1.types import CustomJob, PipelineJob, TrainingPipeline
[docs]class CustomJobHook(GoogleBaseHook):
"""Hook for Google Cloud Vertex AI Custom Job APIs."""
def __init__(
self,
gcp_conn_id: str = "google_cloud_default",
impersonation_chain: str | Sequence[str] | None = None,
**kwargs,
) -> None:
super().__init__(
gcp_conn_id=gcp_conn_id,
impersonation_chain=impersonation_chain,
**kwargs,
)
self._job: None | (
CustomContainerTrainingJob | CustomPythonPackageTrainingJob | CustomTrainingJob
) = None
[docs] def get_pipeline_service_client(
self,
region: str | None = None,
) -> PipelineServiceClient:
"""Return PipelineServiceClient object."""
if region and region != "global":
client_options = ClientOptions(api_endpoint=f"{region}-aiplatform.googleapis.com:443")
else:
client_options = ClientOptions()
return PipelineServiceClient(
credentials=self.get_credentials(), client_info=CLIENT_INFO, client_options=client_options
)
[docs] def get_job_service_client(
self,
region: str | None = None,
) -> JobServiceClient:
"""Return JobServiceClient object."""
if region and region != "global":
client_options = ClientOptions(api_endpoint=f"{region}-aiplatform.googleapis.com:443")
else:
client_options = ClientOptions()
return JobServiceClient(
credentials=self.get_credentials(), client_info=CLIENT_INFO, client_options=client_options
)
[docs] def get_custom_container_training_job(
self,
display_name: str,
container_uri: str,
command: Sequence[str] = (),
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
project: str | None = None,
location: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
) -> CustomContainerTrainingJob:
"""Return CustomContainerTrainingJob object."""
return CustomContainerTrainingJob(
display_name=display_name,
container_uri=container_uri,
command=command,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
project=project,
location=location,
credentials=self.get_credentials(),
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
[docs] def get_custom_python_package_training_job(
self,
display_name: str,
python_package_gcs_uri: str,
python_module_name: str,
container_uri: str,
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
project: str | None = None,
location: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
) -> CustomPythonPackageTrainingJob:
"""Return CustomPythonPackageTrainingJob object."""
return CustomPythonPackageTrainingJob(
display_name=display_name,
container_uri=container_uri,
python_package_gcs_uri=python_package_gcs_uri,
python_module_name=python_module_name,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
project=project,
location=location,
credentials=self.get_credentials(),
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
[docs] def get_custom_training_job(
self,
display_name: str,
script_path: str,
container_uri: str,
requirements: Sequence[str] | None = None,
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
project: str | None = None,
location: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
) -> CustomTrainingJob:
"""Return CustomTrainingJob object."""
return CustomTrainingJob(
display_name=display_name,
script_path=script_path,
container_uri=container_uri,
requirements=requirements,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
project=project,
location=location,
credentials=self.get_credentials(),
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
@staticmethod
@staticmethod
@staticmethod
@staticmethod
@staticmethod
[docs] def wait_for_operation(self, operation: Operation, timeout: float | None = None):
"""Wait for long-lasting operation to complete."""
try:
return operation.result(timeout=timeout)
except Exception:
error = operation.exception(timeout=timeout)
raise AirflowException(error)
[docs] def cancel_job(self) -> None:
"""Cancel Job for training pipeline."""
if self._job:
self._job.cancel()
def _run_job(
self,
job: (CustomTrainingJob | CustomContainerTrainingJob | CustomPythonPackageTrainingJob),
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
sync=True,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
) -> tuple[models.Model | None, str, str]:
"""Run a training pipeline job and wait until its completion."""
model = job.run(
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
sync=sync,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
)
training_id = self.extract_training_id(job.resource_name)
custom_job_id = self.extract_custom_job_id(
job.gca_resource.training_task_metadata.get("backingCustomJob")
)
if model:
model.wait()
else:
self.log.warning(
"Training did not produce a Managed Model returning None. Training Pipeline is not "
"configured to upload a Model. Create the Training Pipeline with "
"model_serving_container_image_uri and model_display_name passed in. "
"Ensure that your training script saves to model to os.environ['AIP_MODEL_DIR']."
)
return model, training_id, custom_job_id
@GoogleBaseHook.fallback_to_default_project_id
@deprecated(
planned_removal_date="March 01, 2025",
use_instead="PipelineJobHook.cancel_pipeline_job",
category=AirflowProviderDeprecationWarning,
)
[docs] def cancel_pipeline_job(
self,
project_id: str,
region: str,
pipeline_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> None:
"""
Cancel a PipelineJob.
Starts asynchronous cancellation on the PipelineJob. The server makes the best
effort to cancel the pipeline, but success is not guaranteed. Clients can use
[PipelineService.GetPipelineJob][google.cloud.aiplatform.v1.PipelineService.GetPipelineJob] or other
methods to check whether the cancellation succeeded or whether the pipeline completed despite
cancellation. On successful cancellation, the PipelineJob is not deleted; instead it becomes a
pipeline with a [PipelineJob.error][google.cloud.aiplatform.v1.PipelineJob.error] value with a
[google.rpc.Status.code][google.rpc.Status.code] of 1, corresponding to ``Code.CANCELLED``, and
[PipelineJob.state][google.cloud.aiplatform.v1.PipelineJob.state] is set to ``CANCELLED``.
This method is deprecated, please use `PipelineJobHook.cancel_pipeline_job` method.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param pipeline_job: The name of the PipelineJob to cancel.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
name = client.pipeline_job_path(project_id, region, pipeline_job)
client.cancel_pipeline_job(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
@GoogleBaseHook.fallback_to_default_project_id
[docs] def cancel_training_pipeline(
self,
project_id: str,
region: str,
training_pipeline: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> None:
"""
Cancel a TrainingPipeline.
Starts asynchronous cancellation on the TrainingPipeline. The server makes
the best effort to cancel the pipeline, but success is not guaranteed. Clients can use
[PipelineService.GetTrainingPipeline][google.cloud.aiplatform.v1.PipelineService.GetTrainingPipeline]
or other methods to check whether the cancellation succeeded or whether the pipeline completed despite
cancellation. On successful cancellation, the TrainingPipeline is not deleted; instead it becomes a
pipeline with a [TrainingPipeline.error][google.cloud.aiplatform.v1.TrainingPipeline.error] value with
a [google.rpc.Status.code][google.rpc.Status.code] of 1, corresponding to ``Code.CANCELLED``, and
[TrainingPipeline.state][google.cloud.aiplatform.v1.TrainingPipeline.state] is set to ``CANCELLED``.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param training_pipeline: Required. The name of the TrainingPipeline to cancel.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
name = client.training_pipeline_path(project_id, region, training_pipeline)
client.cancel_training_pipeline(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
@GoogleBaseHook.fallback_to_default_project_id
[docs] def cancel_custom_job(
self,
project_id: str,
region: str,
custom_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> None:
"""
Cancel a CustomJob.
Starts asynchronous cancellation on the CustomJob. The server makes the best effort
to cancel the job, but success is not guaranteed. Clients can use
[JobService.GetCustomJob][google.cloud.aiplatform.v1.JobService.GetCustomJob] or other methods to
check whether the cancellation succeeded or whether the job completed despite cancellation. On
successful cancellation, the CustomJob is not deleted; instead it becomes a job with a
[CustomJob.error][google.cloud.aiplatform.v1.CustomJob.error] value with a
[google.rpc.Status.code][google.rpc.Status.code] of 1, corresponding to ``Code.CANCELLED``, and
[CustomJob.state][google.cloud.aiplatform.v1.CustomJob.state] is set to ``CANCELLED``.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param custom_job: Required. The name of the CustomJob to cancel.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_job_service_client(region)
name = JobServiceClient.custom_job_path(project_id, region, custom_job)
client.cancel_custom_job(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
@GoogleBaseHook.fallback_to_default_project_id
@deprecated(
planned_removal_date="March 01, 2025",
use_instead="PipelineJobHook.create_pipeline_job",
category=AirflowProviderDeprecationWarning,
)
[docs] def create_pipeline_job(
self,
project_id: str,
region: str,
pipeline_job: PipelineJob,
pipeline_job_id: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> PipelineJob:
"""
Create a PipelineJob. A PipelineJob will run immediately when created.
This method is deprecated, please use `PipelineJobHook.create_pipeline_job` method.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param pipeline_job: Required. The PipelineJob to create.
:param pipeline_job_id: The ID to use for the PipelineJob, which will become the final component of
the PipelineJob name. If not provided, an ID will be automatically generated.
This value should be less than 128 characters, and valid characters are /[a-z][0-9]-/.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
parent = client.common_location_path(project_id, region)
result = client.create_pipeline_job(
request={
"parent": parent,
"pipeline_job": pipeline_job,
"pipeline_job_id": pipeline_job_id,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def create_training_pipeline(
self,
project_id: str,
region: str,
training_pipeline: TrainingPipeline,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> TrainingPipeline:
"""
Create a TrainingPipeline. A created TrainingPipeline right away will be attempted to be run.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param training_pipeline: Required. The TrainingPipeline to create.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
parent = client.common_location_path(project_id, region)
result = client.create_training_pipeline(
request={
"parent": parent,
"training_pipeline": training_pipeline,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def create_custom_job(
self,
project_id: str,
region: str,
custom_job: CustomJob,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> CustomJob:
"""
Create a CustomJob. A created CustomJob right away will be attempted to be run.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param custom_job: Required. The CustomJob to create. This corresponds to the ``custom_job`` field on
the ``request`` instance; if ``request`` is provided, this should not be set.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_job_service_client(region)
parent = JobServiceClient.common_location_path(project_id, region)
result = client.create_custom_job(
request={
"parent": parent,
"custom_job": custom_job,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def create_custom_container_training_job(
self,
project_id: str,
region: str,
display_name: str,
container_uri: str,
command: Sequence[str] = (),
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
# RUN
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
sync=True,
) -> tuple[models.Model | None, str, str]:
"""
Create Custom Container Training Job.
:param display_name: Required. The user-defined name of this TrainingPipeline.
:param command: The command to be invoked when the container is started.
It overrides the entrypoint instruction in Dockerfile when provided
:param container_uri: Required: Uri of the training container image in the GCR.
:param model_serving_container_image_uri: If the training produces a managed Vertex AI Model, the URI
of the Model serving container suitable for serving the model produced by the
training script.
:param model_serving_container_predict_route: If the training produces a managed Vertex AI Model, An
HTTP path to send prediction requests to the container, and which must be supported
by it. If not specified a default HTTP path will be used by Vertex AI.
:param model_serving_container_health_route: If the training produces a managed Vertex AI Model, an
HTTP path to send health check requests to the container, and which must be supported
by it. If not specified a standard HTTP path will be used by AI Platform.
:param model_serving_container_command: The command with which the container is run. Not executed
within a shell. The Docker image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's
environment. If a variable cannot be resolved, the reference in the
input string will be unchanged. The $(VAR_NAME) syntax can be escaped
with a double $$, ie: $$(VAR_NAME). Escaped references will never be
expanded, regardless of whether the variable exists or not.
:param model_serving_container_args: The arguments to the command. The Docker image's CMD is used if
this is not provided. Variable references $(VAR_NAME) are expanded using the
container's environment. If a variable cannot be resolved, the reference
in the input string will be unchanged. The $(VAR_NAME) syntax can be
escaped with a double $$, ie: $$(VAR_NAME). Escaped references will
never be expanded, regardless of whether the variable exists or not.
:param model_serving_container_environment_variables: The environment variables that are to be
present in the container. Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
:param model_serving_container_ports: Declaration of ports that are exposed by the container. This
field is primarily informational, it gives Vertex AI information about the
network connections the container uses. Listing or not a port here has
no impact on whether the port is actually exposed, any port listening on
the default "0.0.0.0" address inside a container will be accessible from
the network.
:param model_description: The description of the Model.
:param model_instance_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single instance, which
are used in
``PredictRequest.instances``,
``ExplainRequest.instances``
and
``BatchPredictionJob.input_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param model_parameters_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the parameters of prediction and
explanation via
``PredictRequest.parameters``,
``ExplainRequest.parameters``
and
``BatchPredictionJob.model_parameters``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform, if no parameters are supported it is set to an
empty string. Note: The URI given on output will be
immutable and probably different, including the URI scheme,
than the one given on input. The output URI will point to a
location where the user only has a read access.
:param model_prediction_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single prediction
produced by this Model, which are returned via
``PredictResponse.predictions``,
``ExplainResponse.explanations``,
and
``BatchPredictionJob.output_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param parent_model: Optional. The resource name or model ID of an existing model.
The new model uploaded by this job will be a version of `parent_model`.
Only set this field when training a new version of an existing model.
:param is_default_version: Optional. When set to True, the newly uploaded model version will
automatically have alias "default" included. Subsequent uses of
the model produced by this job without a version specified will
use this "default" version.
When set to False, the "default" alias will not be moved.
Actions targeting the model version produced by this job will need
to specifically reference this version by ID or alias.
New model uploads, i.e. version 1, will always be "default" aliased.
:param model_version_aliases: Optional. User provided version aliases so that the model version
uploaded by this job can be referenced via alias instead of
auto-generated version ID. A default version alias will be created
for the first version of the model.
The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9]
:param model_version_description: Optional. The description of the model version
being uploaded by this job.
:param project_id: Project to run training in.
:param region: Location to run training in.
:param labels: Optional. The labels with user-defined metadata to
organize TrainingPipelines.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param training_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the training pipeline. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, this TrainingPipeline will be secured by this key.
Note: Model trained by this TrainingPipeline is also secured
by this key if ``model_to_upload`` is not set separately.
:param model_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the model. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, the trained Model will be secured by this key.
:param staging_bucket: Bucket used to stage source and training artifacts.
:param dataset: Vertex AI to fit this training against.
:param annotation_schema_uri: Google Cloud Storage URI points to a YAML file describing
annotation schema. The schema is defined as an OpenAPI 3.0.2
[Schema Object]
(https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schema-object)
Only Annotations that both match this schema and belong to
DataItems not ignored by the split method are used in
respectively training, validation or test role, depending on
the role of the DataItem they are on.
When used in conjunction with
``annotations_filter``,
the Annotations used for training are filtered by both
``annotations_filter``
and
``annotation_schema_uri``.
:param model_display_name: If the script produces a managed Vertex AI Model. The display name of
the Model. The name can be up to 128 characters long and can be consist
of any UTF-8 characters.
If not provided upon creation, the job's display_name is used.
:param model_labels: Optional. The labels with user-defined metadata to
organize your Models.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param base_output_dir: GCS output directory of job. If not provided a timestamped directory in the
staging directory will be used.
Vertex AI sets the following environment variables when it runs your training code:
- AIP_MODEL_DIR: a Cloud Storage URI of a directory intended for saving model artifacts,
i.e. <base_output_dir>/model/
- AIP_CHECKPOINT_DIR: a Cloud Storage URI of a directory intended for saving checkpoints,
i.e. <base_output_dir>/checkpoints/
- AIP_TENSORBOARD_LOG_DIR: a Cloud Storage URI of a directory intended for saving TensorBoard
logs, i.e. <base_output_dir>/logs/
:param service_account: Specifies the service account for workload run-as account.
Users submitting jobs must have act-as permission on this run-as account.
:param network: The full name of the Compute Engine network to which the job
should be peered.
Private services access must already be configured for the network.
If left unspecified, the job is not peered with any network.
:param bigquery_destination: Provide this field if `dataset` is a BiqQuery dataset.
The BigQuery project location where the training data is to
be written to. In the given project a new dataset is created
with name
``dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>``
where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All
training input data will be written into that dataset. In
the dataset three tables will be created, ``training``,
``validation`` and ``test``.
- AIP_DATA_FORMAT = "bigquery".
- AIP_TRAINING_DATA_URI ="bigquery_destination.dataset_*.training"
- AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_*.validation"
- AIP_TEST_DATA_URI = "bigquery_destination.dataset_*.test"
:param args: Command line arguments to be passed to the Python script.
:param environment_variables: Environment variables to be passed to the container.
Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
At most 10 environment variables can be specified.
The Name of the environment variable must be unique.
:param replica_count: The number of worker replicas. If replica count = 1 then one chief
replica will be provisioned. If replica_count > 1 the remainder will be
provisioned as a worker replica pool.
:param machine_type: The type of machine to use for training.
:param accelerator_type: Hardware accelerator type. One of ACCELERATOR_TYPE_UNSPECIFIED,
NVIDIA_TESLA_K80, NVIDIA_TESLA_P100, NVIDIA_TESLA_V100, NVIDIA_TESLA_P4,
NVIDIA_TESLA_T4
:param accelerator_count: The number of accelerators to attach to a worker replica.
:param 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).
:param boot_disk_size_gb: Size in GB of the boot disk, default is 100GB.
boot disk size must be within the range of [100, 64000].
:param training_fraction_split: Optional. The fraction of the input data that is to be used to train
the Model. This is ignored if Dataset is not provided.
:param validation_fraction_split: Optional. The fraction of the input data that is to be used to
validate the Model. This is ignored if Dataset is not provided.
:param test_fraction_split: Optional. The fraction of the input data that is to be used to evaluate
the Model. This is ignored if Dataset is not provided.
:param training_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to train the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param validation_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to validate the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param test_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to test the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param predefined_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key (either the label's value or
value in the column) must be one of {``training``,
``validation``, ``test``}, and it defines to which set the
given piece of data is assigned. If for a piece of data the
key is not present or has an invalid value, that piece is
ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param timestamp_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key values of the key (the values in
the column) must be in RFC 3339 `date-time` format, where
`time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a
piece of data the key is not present or has an invalid value,
that piece is ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param tensorboard: Optional. The name of a Vertex AI resource to which this CustomJob will upload
logs. Format:
``projects/{project}/locations/{location}/tensorboards/{tensorboard}``
For more information on configuring your service account please visit:
https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-training
:param sync: Whether to execute the AI Platform job synchronously. If False, this method
will be executed in concurrent Future and any downstream object will
be immediately returned and synced when the Future has completed.
"""
self._job = self.get_custom_container_training_job(
project=project_id,
location=region,
display_name=display_name,
container_uri=container_uri,
command=command,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
if not self._job:
raise AirflowException("CustomJob was not created")
model, training_id, custom_job_id = self._run_job(
job=self._job,
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
sync=sync,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
)
return model, training_id, custom_job_id
@GoogleBaseHook.fallback_to_default_project_id
[docs] def create_custom_python_package_training_job(
self,
project_id: str,
region: str,
display_name: str,
python_package_gcs_uri: str,
python_module_name: str,
container_uri: str,
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
# RUN
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
sync=True,
) -> tuple[models.Model | None, str, str]:
"""
Create Custom Python Package Training Job.
:param display_name: Required. The user-defined name of this TrainingPipeline.
:param python_package_gcs_uri: Required: GCS location of the training python package.
:param python_module_name: Required: The module name of the training python package.
:param container_uri: Required: Uri of the training container image in the GCR.
:param model_serving_container_image_uri: If the training produces a managed Vertex AI Model, the URI
of the Model serving container suitable for serving the model produced by the
training script.
:param model_serving_container_predict_route: If the training produces a managed Vertex AI Model, An
HTTP path to send prediction requests to the container, and which must be supported
by it. If not specified a default HTTP path will be used by Vertex AI.
:param model_serving_container_health_route: If the training produces a managed Vertex AI Model, an
HTTP path to send health check requests to the container, and which must be supported
by it. If not specified a standard HTTP path will be used by AI Platform.
:param model_serving_container_command: The command with which the container is run. Not executed
within a shell. The Docker image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's
environment. If a variable cannot be resolved, the reference in the
input string will be unchanged. The $(VAR_NAME) syntax can be escaped
with a double $$, ie: $$(VAR_NAME). Escaped references will never be
expanded, regardless of whether the variable exists or not.
:param model_serving_container_args: The arguments to the command. The Docker image's CMD is used if
this is not provided. Variable references $(VAR_NAME) are expanded using the
container's environment. If a variable cannot be resolved, the reference
in the input string will be unchanged. The $(VAR_NAME) syntax can be
escaped with a double $$, ie: $$(VAR_NAME). Escaped references will
never be expanded, regardless of whether the variable exists or not.
:param model_serving_container_environment_variables: The environment variables that are to be
present in the container. Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
:param model_serving_container_ports: Declaration of ports that are exposed by the container. This
field is primarily informational, it gives Vertex AI information about the
network connections the container uses. Listing or not a port here has
no impact on whether the port is actually exposed, any port listening on
the default "0.0.0.0" address inside a container will be accessible from
the network.
:param model_description: The description of the Model.
:param model_instance_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single instance, which
are used in
``PredictRequest.instances``,
``ExplainRequest.instances``
and
``BatchPredictionJob.input_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param model_parameters_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the parameters of prediction and
explanation via
``PredictRequest.parameters``,
``ExplainRequest.parameters``
and
``BatchPredictionJob.model_parameters``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform, if no parameters are supported it is set to an
empty string. Note: The URI given on output will be
immutable and probably different, including the URI scheme,
than the one given on input. The output URI will point to a
location where the user only has a read access.
:param model_prediction_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single prediction
produced by this Model, which are returned via
``PredictResponse.predictions``,
``ExplainResponse.explanations``,
and
``BatchPredictionJob.output_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param parent_model: Optional. The resource name or model ID of an existing model.
The new model uploaded by this job will be a version of `parent_model`.
Only set this field when training a new version of an existing model.
:param is_default_version: Optional. When set to True, the newly uploaded model version will
automatically have alias "default" included. Subsequent uses of
the model produced by this job without a version specified will
use this "default" version.
When set to False, the "default" alias will not be moved.
Actions targeting the model version produced by this job will need
to specifically reference this version by ID or alias.
New model uploads, i.e. version 1, will always be "default" aliased.
:param model_version_aliases: Optional. User provided version aliases so that the model version
uploaded by this job can be referenced via alias instead of
auto-generated version ID. A default version alias will be created
for the first version of the model.
The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9]
:param model_version_description: Optional. The description of the model version
being uploaded by this job.
:param project_id: Project to run training in.
:param region: Location to run training in.
:param labels: Optional. The labels with user-defined metadata to
organize TrainingPipelines.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param training_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the training pipeline. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, this TrainingPipeline will be secured by this key.
Note: Model trained by this TrainingPipeline is also secured
by this key if ``model_to_upload`` is not set separately.
:param model_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the model. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, the trained Model will be secured by this key.
:param staging_bucket: Bucket used to stage source and training artifacts.
:param dataset: Vertex AI to fit this training against.
:param annotation_schema_uri: Google Cloud Storage URI points to a YAML file describing
annotation schema. The schema is defined as an OpenAPI 3.0.2
[Schema Object]
(https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schema-object)
Only Annotations that both match this schema and belong to
DataItems not ignored by the split method are used in
respectively training, validation or test role, depending on
the role of the DataItem they are on.
When used in conjunction with
``annotations_filter``,
the Annotations used for training are filtered by both
``annotations_filter``
and
``annotation_schema_uri``.
:param model_display_name: If the script produces a managed Vertex AI Model. The display name of
the Model. The name can be up to 128 characters long and can be consist
of any UTF-8 characters.
If not provided upon creation, the job's display_name is used.
:param model_labels: Optional. The labels with user-defined metadata to
organize your Models.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param base_output_dir: GCS output directory of job. If not provided a timestamped directory in the
staging directory will be used.
Vertex AI sets the following environment variables when it runs your training code:
- AIP_MODEL_DIR: a Cloud Storage URI of a directory intended for saving model artifacts,
i.e. <base_output_dir>/model/
- AIP_CHECKPOINT_DIR: a Cloud Storage URI of a directory intended for saving checkpoints,
i.e. <base_output_dir>/checkpoints/
- AIP_TENSORBOARD_LOG_DIR: a Cloud Storage URI of a directory intended for saving TensorBoard
logs, i.e. <base_output_dir>/logs/
:param service_account: Specifies the service account for workload run-as account.
Users submitting jobs must have act-as permission on this run-as account.
:param network: The full name of the Compute Engine network to which the job
should be peered.
Private services access must already be configured for the network.
If left unspecified, the job is not peered with any network.
:param bigquery_destination: Provide this field if `dataset` is a BiqQuery dataset.
The BigQuery project location where the training data is to
be written to. In the given project a new dataset is created
with name
``dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>``
where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All
training input data will be written into that dataset. In
the dataset three tables will be created, ``training``,
``validation`` and ``test``.
- AIP_DATA_FORMAT = "bigquery".
- AIP_TRAINING_DATA_URI ="bigquery_destination.dataset_*.training"
- AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_*.validation"
- AIP_TEST_DATA_URI = "bigquery_destination.dataset_*.test"
:param args: Command line arguments to be passed to the Python script.
:param environment_variables: Environment variables to be passed to the container.
Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
At most 10 environment variables can be specified.
The Name of the environment variable must be unique.
:param replica_count: The number of worker replicas. If replica count = 1 then one chief
replica will be provisioned. If replica_count > 1 the remainder will be
provisioned as a worker replica pool.
:param machine_type: The type of machine to use for training.
:param accelerator_type: Hardware accelerator type. One of ACCELERATOR_TYPE_UNSPECIFIED,
NVIDIA_TESLA_K80, NVIDIA_TESLA_P100, NVIDIA_TESLA_V100, NVIDIA_TESLA_P4,
NVIDIA_TESLA_T4
:param accelerator_count: The number of accelerators to attach to a worker replica.
:param 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).
:param boot_disk_size_gb: Size in GB of the boot disk, default is 100GB.
boot disk size must be within the range of [100, 64000].
:param training_fraction_split: Optional. The fraction of the input data that is to be used to train
the Model. This is ignored if Dataset is not provided.
:param validation_fraction_split: Optional. The fraction of the input data that is to be used to
validate the Model. This is ignored if Dataset is not provided.
:param test_fraction_split: Optional. The fraction of the input data that is to be used to evaluate
the Model. This is ignored if Dataset is not provided.
:param training_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to train the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param validation_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to validate the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param test_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to test the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param predefined_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key (either the label's value or
value in the column) must be one of {``training``,
``validation``, ``test``}, and it defines to which set the
given piece of data is assigned. If for a piece of data the
key is not present or has an invalid value, that piece is
ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param timestamp_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key values of the key (the values in
the column) must be in RFC 3339 `date-time` format, where
`time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a
piece of data the key is not present or has an invalid value,
that piece is ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param tensorboard: Optional. The name of a Vertex AI resource to which this CustomJob will upload
logs. Format:
``projects/{project}/locations/{location}/tensorboards/{tensorboard}``
For more information on configuring your service account please visit:
https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-training
:param sync: Whether to execute the AI Platform job synchronously. If False, this method
will be executed in concurrent Future and any downstream object will
be immediately returned and synced when the Future has completed.
"""
self._job = self.get_custom_python_package_training_job(
project=project_id,
location=region,
display_name=display_name,
python_package_gcs_uri=python_package_gcs_uri,
python_module_name=python_module_name,
container_uri=container_uri,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
if not self._job:
raise AirflowException("CustomJob was not created")
model, training_id, custom_job_id = self._run_job(
job=self._job,
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
sync=sync,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
)
return model, training_id, custom_job_id
@GoogleBaseHook.fallback_to_default_project_id
[docs] def create_custom_training_job(
self,
project_id: str,
region: str,
display_name: str,
script_path: str,
container_uri: str,
requirements: Sequence[str] | None = None,
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
# RUN
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
sync=True,
) -> tuple[models.Model | None, str, str]:
"""
Create Custom Training Job.
:param display_name: Required. The user-defined name of this TrainingPipeline.
:param script_path: Required. Local path to training script.
:param container_uri: Required: Uri of the training container image in the GCR.
:param requirements: List of python packages dependencies of script.
:param model_serving_container_image_uri: If the training produces a managed Vertex AI Model, the URI
of the Model serving container suitable for serving the model produced by the
training script.
:param model_serving_container_predict_route: If the training produces a managed Vertex AI Model, An
HTTP path to send prediction requests to the container, and which must be supported
by it. If not specified a default HTTP path will be used by Vertex AI.
:param model_serving_container_health_route: If the training produces a managed Vertex AI Model, an
HTTP path to send health check requests to the container, and which must be supported
by it. If not specified a standard HTTP path will be used by AI Platform.
:param model_serving_container_command: The command with which the container is run. Not executed
within a shell. The Docker image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's
environment. If a variable cannot be resolved, the reference in the
input string will be unchanged. The $(VAR_NAME) syntax can be escaped
with a double $$, ie: $$(VAR_NAME). Escaped references will never be
expanded, regardless of whether the variable exists or not.
:param model_serving_container_args: The arguments to the command. The Docker image's CMD is used if
this is not provided. Variable references $(VAR_NAME) are expanded using the
container's environment. If a variable cannot be resolved, the reference
in the input string will be unchanged. The $(VAR_NAME) syntax can be
escaped with a double $$, ie: $$(VAR_NAME). Escaped references will
never be expanded, regardless of whether the variable exists or not.
:param model_serving_container_environment_variables: The environment variables that are to be
present in the container. Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
:param model_serving_container_ports: Declaration of ports that are exposed by the container. This
field is primarily informational, it gives Vertex AI information about the
network connections the container uses. Listing or not a port here has
no impact on whether the port is actually exposed, any port listening on
the default "0.0.0.0" address inside a container will be accessible from
the network.
:param model_description: The description of the Model.
:param model_instance_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single instance, which
are used in
``PredictRequest.instances``,
``ExplainRequest.instances``
and
``BatchPredictionJob.input_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param model_parameters_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the parameters of prediction and
explanation via
``PredictRequest.parameters``,
``ExplainRequest.parameters``
and
``BatchPredictionJob.model_parameters``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform, if no parameters are supported it is set to an
empty string. Note: The URI given on output will be
immutable and probably different, including the URI scheme,
than the one given on input. The output URI will point to a
location where the user only has a read access.
:param model_prediction_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single prediction
produced by this Model, which are returned via
``PredictResponse.predictions``,
``ExplainResponse.explanations``,
and
``BatchPredictionJob.output_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param parent_model: Optional. The resource name or model ID of an existing model.
The new model uploaded by this job will be a version of `parent_model`.
Only set this field when training a new version of an existing model.
:param is_default_version: Optional. When set to True, the newly uploaded model version will
automatically have alias "default" included. Subsequent uses of
the model produced by this job without a version specified will
use this "default" version.
When set to False, the "default" alias will not be moved.
Actions targeting the model version produced by this job will need
to specifically reference this version by ID or alias.
New model uploads, i.e. version 1, will always be "default" aliased.
:param model_version_aliases: Optional. User provided version aliases so that the model version
uploaded by this job can be referenced via alias instead of
auto-generated version ID. A default version alias will be created
for the first version of the model.
The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9]
:param model_version_description: Optional. The description of the model version
being uploaded by this job.
:param project_id: Project to run training in.
:param region: Location to run training in.
:param labels: Optional. The labels with user-defined metadata to
organize TrainingPipelines.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param training_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the training pipeline. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, this TrainingPipeline will be secured by this key.
Note: Model trained by this TrainingPipeline is also secured
by this key if ``model_to_upload`` is not set separately.
:param model_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the model. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, the trained Model will be secured by this key.
:param staging_bucket: Bucket used to stage source and training artifacts.
:param dataset: Vertex AI to fit this training against.
:param annotation_schema_uri: Google Cloud Storage URI points to a YAML file describing
annotation schema. The schema is defined as an OpenAPI 3.0.2
[Schema Object]
(https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schema-object)
Only Annotations that both match this schema and belong to
DataItems not ignored by the split method are used in
respectively training, validation or test role, depending on
the role of the DataItem they are on.
When used in conjunction with
``annotations_filter``,
the Annotations used for training are filtered by both
``annotations_filter``
and
``annotation_schema_uri``.
:param model_display_name: If the script produces a managed Vertex AI Model. The display name of
the Model. The name can be up to 128 characters long and can be consist
of any UTF-8 characters.
If not provided upon creation, the job's display_name is used.
:param model_labels: Optional. The labels with user-defined metadata to
organize your Models.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param base_output_dir: GCS output directory of job. If not provided a timestamped directory in the
staging directory will be used.
Vertex AI sets the following environment variables when it runs your training code:
- AIP_MODEL_DIR: a Cloud Storage URI of a directory intended for saving model artifacts,
i.e. <base_output_dir>/model/
- AIP_CHECKPOINT_DIR: a Cloud Storage URI of a directory intended for saving checkpoints,
i.e. <base_output_dir>/checkpoints/
- AIP_TENSORBOARD_LOG_DIR: a Cloud Storage URI of a directory intended for saving TensorBoard
logs, i.e. <base_output_dir>/logs/
:param service_account: Specifies the service account for workload run-as account.
Users submitting jobs must have act-as permission on this run-as account.
:param network: The full name of the Compute Engine network to which the job
should be peered.
Private services access must already be configured for the network.
If left unspecified, the job is not peered with any network.
:param bigquery_destination: Provide this field if `dataset` is a BiqQuery dataset.
The BigQuery project location where the training data is to
be written to. In the given project a new dataset is created
with name
``dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>``
where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All
training input data will be written into that dataset. In
the dataset three tables will be created, ``training``,
``validation`` and ``test``.
- AIP_DATA_FORMAT = "bigquery".
- AIP_TRAINING_DATA_URI ="bigquery_destination.dataset_*.training"
- AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_*.validation"
- AIP_TEST_DATA_URI = "bigquery_destination.dataset_*.test"
:param args: Command line arguments to be passed to the Python script.
:param environment_variables: Environment variables to be passed to the container.
Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
At most 10 environment variables can be specified.
The Name of the environment variable must be unique.
:param replica_count: The number of worker replicas. If replica count = 1 then one chief
replica will be provisioned. If replica_count > 1 the remainder will be
provisioned as a worker replica pool.
:param machine_type: The type of machine to use for training.
:param accelerator_type: Hardware accelerator type. One of ACCELERATOR_TYPE_UNSPECIFIED,
NVIDIA_TESLA_K80, NVIDIA_TESLA_P100, NVIDIA_TESLA_V100, NVIDIA_TESLA_P4,
NVIDIA_TESLA_T4
:param accelerator_count: The number of accelerators to attach to a worker replica.
:param 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).
:param boot_disk_size_gb: Size in GB of the boot disk, default is 100GB.
boot disk size must be within the range of [100, 64000].
:param training_fraction_split: Optional. The fraction of the input data that is to be used to train
the Model. This is ignored if Dataset is not provided.
:param validation_fraction_split: Optional. The fraction of the input data that is to be used to
validate the Model. This is ignored if Dataset is not provided.
:param test_fraction_split: Optional. The fraction of the input data that is to be used to evaluate
the Model. This is ignored if Dataset is not provided.
:param training_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to train the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param validation_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to validate the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param test_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to test the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param predefined_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key (either the label's value or
value in the column) must be one of {``training``,
``validation``, ``test``}, and it defines to which set the
given piece of data is assigned. If for a piece of data the
key is not present or has an invalid value, that piece is
ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param timestamp_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key values of the key (the values in
the column) must be in RFC 3339 `date-time` format, where
`time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a
piece of data the key is not present or has an invalid value,
that piece is ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param tensorboard: Optional. The name of a Vertex AI resource to which this CustomJob will upload
logs. Format:
``projects/{project}/locations/{location}/tensorboards/{tensorboard}``
For more information on configuring your service account please visit:
https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-training
:param sync: Whether to execute the AI Platform job synchronously. If False, this method
will be executed in concurrent Future and any downstream object will
be immediately returned and synced when the Future has completed.
"""
self._job = self.get_custom_training_job(
project=project_id,
location=region,
display_name=display_name,
script_path=script_path,
container_uri=container_uri,
requirements=requirements,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
if not self._job:
raise AirflowException("CustomJob was not created")
model, training_id, custom_job_id = self._run_job(
job=self._job,
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
sync=sync,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
)
return model, training_id, custom_job_id
@GoogleBaseHook.fallback_to_default_project_id
[docs] def submit_custom_container_training_job(
self,
*,
project_id: str,
region: str,
display_name: str,
container_uri: str,
command: Sequence[str] = (),
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
# RUN
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
) -> CustomContainerTrainingJob:
"""
Create and submit a Custom Container Training Job pipeline, then exit without waiting for it to complete.
:param display_name: Required. The user-defined name of this TrainingPipeline.
:param command: The command to be invoked when the container is started.
It overrides the entrypoint instruction in Dockerfile when provided
:param container_uri: Required: Uri of the training container image in the GCR.
:param model_serving_container_image_uri: If the training produces a managed Vertex AI Model, the URI
of the Model serving container suitable for serving the model produced by the
training script.
:param model_serving_container_predict_route: If the training produces a managed Vertex AI Model, An
HTTP path to send prediction requests to the container, and which must be supported
by it. If not specified a default HTTP path will be used by Vertex AI.
:param model_serving_container_health_route: If the training produces a managed Vertex AI Model, an
HTTP path to send health check requests to the container, and which must be supported
by it. If not specified a standard HTTP path will be used by AI Platform.
:param model_serving_container_command: The command with which the container is run. Not executed
within a shell. The Docker image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's
environment. If a variable cannot be resolved, the reference in the
input string will be unchanged. The $(VAR_NAME) syntax can be escaped
with a double $$, ie: $$(VAR_NAME). Escaped references will never be
expanded, regardless of whether the variable exists or not.
:param model_serving_container_args: The arguments to the command. The Docker image's CMD is used if
this is not provided. Variable references $(VAR_NAME) are expanded using the
container's environment. If a variable cannot be resolved, the reference
in the input string will be unchanged. The $(VAR_NAME) syntax can be
escaped with a double $$, ie: $$(VAR_NAME). Escaped references will
never be expanded, regardless of whether the variable exists or not.
:param model_serving_container_environment_variables: The environment variables that are to be
present in the container. Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
:param model_serving_container_ports: Declaration of ports that are exposed by the container. This
field is primarily informational, it gives Vertex AI information about the
network connections the container uses. Listing or not a port here has
no impact on whether the port is actually exposed, any port listening on
the default "0.0.0.0" address inside a container will be accessible from
the network.
:param model_description: The description of the Model.
:param model_instance_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single instance, which
are used in
``PredictRequest.instances``,
``ExplainRequest.instances``
and
``BatchPredictionJob.input_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param model_parameters_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the parameters of prediction and
explanation via
``PredictRequest.parameters``,
``ExplainRequest.parameters``
and
``BatchPredictionJob.model_parameters``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform, if no parameters are supported it is set to an
empty string. Note: The URI given on output will be
immutable and probably different, including the URI scheme,
than the one given on input. The output URI will point to a
location where the user only has a read access.
:param model_prediction_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single prediction
produced by this Model, which are returned via
``PredictResponse.predictions``,
``ExplainResponse.explanations``,
and
``BatchPredictionJob.output_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param parent_model: Optional. The resource name or model ID of an existing model.
The new model uploaded by this job will be a version of `parent_model`.
Only set this field when training a new version of an existing model.
:param is_default_version: Optional. When set to True, the newly uploaded model version will
automatically have alias "default" included. Subsequent uses of
the model produced by this job without a version specified will
use this "default" version.
When set to False, the "default" alias will not be moved.
Actions targeting the model version produced by this job will need
to specifically reference this version by ID or alias.
New model uploads, i.e. version 1, will always be "default" aliased.
:param model_version_aliases: Optional. User provided version aliases so that the model version
uploaded by this job can be referenced via alias instead of
auto-generated version ID. A default version alias will be created
for the first version of the model.
The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9]
:param model_version_description: Optional. The description of the model version
being uploaded by this job.
:param project_id: Project to run training in.
:param region: Location to run training in.
:param labels: Optional. The labels with user-defined metadata to
organize TrainingPipelines.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param training_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the training pipeline. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, this TrainingPipeline will be secured by this key.
Note: Model trained by this TrainingPipeline is also secured
by this key if ``model_to_upload`` is not set separately.
:param model_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the model. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, the trained Model will be secured by this key.
:param staging_bucket: Bucket used to stage source and training artifacts.
:param dataset: Vertex AI to fit this training against.
:param annotation_schema_uri: Google Cloud Storage URI points to a YAML file describing
annotation schema. The schema is defined as an OpenAPI 3.0.2
[Schema Object]
(https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schema-object)
Only Annotations that both match this schema and belong to
DataItems not ignored by the split method are used in
respectively training, validation or test role, depending on
the role of the DataItem they are on.
When used in conjunction with
``annotations_filter``,
the Annotations used for training are filtered by both
``annotations_filter``
and
``annotation_schema_uri``.
:param model_display_name: If the script produces a managed Vertex AI Model. The display name of
the Model. The name can be up to 128 characters long and can be consist
of any UTF-8 characters.
If not provided upon creation, the job's display_name is used.
:param model_labels: Optional. The labels with user-defined metadata to
organize your Models.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param base_output_dir: GCS output directory of job. If not provided a timestamped directory in the
staging directory will be used.
Vertex AI sets the following environment variables when it runs your training code:
- AIP_MODEL_DIR: a Cloud Storage URI of a directory intended for saving model artifacts,
i.e. <base_output_dir>/model/
- AIP_CHECKPOINT_DIR: a Cloud Storage URI of a directory intended for saving checkpoints,
i.e. <base_output_dir>/checkpoints/
- AIP_TENSORBOARD_LOG_DIR: a Cloud Storage URI of a directory intended for saving TensorBoard
logs, i.e. <base_output_dir>/logs/
:param service_account: Specifies the service account for workload run-as account.
Users submitting jobs must have act-as permission on this run-as account.
:param network: The full name of the Compute Engine network to which the job
should be peered.
Private services access must already be configured for the network.
If left unspecified, the job is not peered with any network.
:param bigquery_destination: Provide this field if `dataset` is a BiqQuery dataset.
The BigQuery project location where the training data is to
be written to. In the given project a new dataset is created
with name
``dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>``
where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All
training input data will be written into that dataset. In
the dataset three tables will be created, ``training``,
``validation`` and ``test``.
- AIP_DATA_FORMAT = "bigquery".
- AIP_TRAINING_DATA_URI ="bigquery_destination.dataset_*.training"
- AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_*.validation"
- AIP_TEST_DATA_URI = "bigquery_destination.dataset_*.test"
:param args: Command line arguments to be passed to the Python script.
:param environment_variables: Environment variables to be passed to the container.
Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
At most 10 environment variables can be specified.
The Name of the environment variable must be unique.
:param replica_count: The number of worker replicas. If replica count = 1 then one chief
replica will be provisioned. If replica_count > 1 the remainder will be
provisioned as a worker replica pool.
:param machine_type: The type of machine to use for training.
:param accelerator_type: Hardware accelerator type. One of ACCELERATOR_TYPE_UNSPECIFIED,
NVIDIA_TESLA_K80, NVIDIA_TESLA_P100, NVIDIA_TESLA_V100, NVIDIA_TESLA_P4,
NVIDIA_TESLA_T4
:param accelerator_count: The number of accelerators to attach to a worker replica.
:param 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).
:param boot_disk_size_gb: Size in GB of the boot disk, default is 100GB.
boot disk size must be within the range of [100, 64000].
:param training_fraction_split: Optional. The fraction of the input data that is to be used to train
the Model. This is ignored if Dataset is not provided.
:param validation_fraction_split: Optional. The fraction of the input data that is to be used to
validate the Model. This is ignored if Dataset is not provided.
:param test_fraction_split: Optional. The fraction of the input data that is to be used to evaluate
the Model. This is ignored if Dataset is not provided.
:param training_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to train the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param validation_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to validate the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param test_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to test the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param predefined_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key (either the label's value or
value in the column) must be one of {``training``,
``validation``, ``test``}, and it defines to which set the
given piece of data is assigned. If for a piece of data the
key is not present or has an invalid value, that piece is
ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param timestamp_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key values of the key (the values in
the column) must be in RFC 3339 `date-time` format, where
`time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a
piece of data the key is not present or has an invalid value,
that piece is ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param tensorboard: Optional. The name of a Vertex AI resource to which this CustomJob will upload
logs. Format:
``projects/{project}/locations/{location}/tensorboards/{tensorboard}``
For more information on configuring your service account please visit:
https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-training
"""
self._job = self.get_custom_container_training_job(
project=project_id,
location=region,
display_name=display_name,
container_uri=container_uri,
command=command,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
if not self._job:
raise AirflowException("CustomContainerTrainingJob instance creation failed.")
self._job.submit(
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
sync=False,
)
return self._job
@GoogleBaseHook.fallback_to_default_project_id
[docs] def submit_custom_python_package_training_job(
self,
*,
project_id: str,
region: str,
display_name: str,
python_package_gcs_uri: str,
python_module_name: str,
container_uri: str,
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
# RUN
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
) -> CustomPythonPackageTrainingJob:
"""
Create and submit a Custom Python Package Training Job pipeline, then exit without waiting for it to complete.
:param display_name: Required. The user-defined name of this TrainingPipeline.
:param python_package_gcs_uri: Required: GCS location of the training python package.
:param python_module_name: Required: The module name of the training python package.
:param container_uri: Required: Uri of the training container image in the GCR.
:param model_serving_container_image_uri: If the training produces a managed Vertex AI Model, the URI
of the Model serving container suitable for serving the model produced by the
training script.
:param model_serving_container_predict_route: If the training produces a managed Vertex AI Model, An
HTTP path to send prediction requests to the container, and which must be supported
by it. If not specified a default HTTP path will be used by Vertex AI.
:param model_serving_container_health_route: If the training produces a managed Vertex AI Model, an
HTTP path to send health check requests to the container, and which must be supported
by it. If not specified a standard HTTP path will be used by AI Platform.
:param model_serving_container_command: The command with which the container is run. Not executed
within a shell. The Docker image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's
environment. If a variable cannot be resolved, the reference in the
input string will be unchanged. The $(VAR_NAME) syntax can be escaped
with a double $$, ie: $$(VAR_NAME). Escaped references will never be
expanded, regardless of whether the variable exists or not.
:param model_serving_container_args: The arguments to the command. The Docker image's CMD is used if
this is not provided. Variable references $(VAR_NAME) are expanded using the
container's environment. If a variable cannot be resolved, the reference
in the input string will be unchanged. The $(VAR_NAME) syntax can be
escaped with a double $$, ie: $$(VAR_NAME). Escaped references will
never be expanded, regardless of whether the variable exists or not.
:param model_serving_container_environment_variables: The environment variables that are to be
present in the container. Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
:param model_serving_container_ports: Declaration of ports that are exposed by the container. This
field is primarily informational, it gives Vertex AI information about the
network connections the container uses. Listing or not a port here has
no impact on whether the port is actually exposed, any port listening on
the default "0.0.0.0" address inside a container will be accessible from
the network.
:param model_description: The description of the Model.
:param model_instance_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single instance, which
are used in
``PredictRequest.instances``,
``ExplainRequest.instances``
and
``BatchPredictionJob.input_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param model_parameters_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the parameters of prediction and
explanation via
``PredictRequest.parameters``,
``ExplainRequest.parameters``
and
``BatchPredictionJob.model_parameters``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform, if no parameters are supported it is set to an
empty string. Note: The URI given on output will be
immutable and probably different, including the URI scheme,
than the one given on input. The output URI will point to a
location where the user only has a read access.
:param model_prediction_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single prediction
produced by this Model, which are returned via
``PredictResponse.predictions``,
``ExplainResponse.explanations``,
and
``BatchPredictionJob.output_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param parent_model: Optional. The resource name or model ID of an existing model.
The new model uploaded by this job will be a version of `parent_model`.
Only set this field when training a new version of an existing model.
:param is_default_version: Optional. When set to True, the newly uploaded model version will
automatically have alias "default" included. Subsequent uses of
the model produced by this job without a version specified will
use this "default" version.
When set to False, the "default" alias will not be moved.
Actions targeting the model version produced by this job will need
to specifically reference this version by ID or alias.
New model uploads, i.e. version 1, will always be "default" aliased.
:param model_version_aliases: Optional. User provided version aliases so that the model version
uploaded by this job can be referenced via alias instead of
auto-generated version ID. A default version alias will be created
for the first version of the model.
The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9]
:param model_version_description: Optional. The description of the model version
being uploaded by this job.
:param project_id: Project to run training in.
:param region: Location to run training in.
:param labels: Optional. The labels with user-defined metadata to
organize TrainingPipelines.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param training_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the training pipeline. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, this TrainingPipeline will be secured by this key.
Note: Model trained by this TrainingPipeline is also secured
by this key if ``model_to_upload`` is not set separately.
:param model_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the model. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, the trained Model will be secured by this key.
:param staging_bucket: Bucket used to stage source and training artifacts.
:param dataset: Vertex AI to fit this training against.
:param annotation_schema_uri: Google Cloud Storage URI points to a YAML file describing
annotation schema. The schema is defined as an OpenAPI 3.0.2
[Schema Object]
(https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schema-object)
Only Annotations that both match this schema and belong to
DataItems not ignored by the split method are used in
respectively training, validation or test role, depending on
the role of the DataItem they are on.
When used in conjunction with
``annotations_filter``,
the Annotations used for training are filtered by both
``annotations_filter``
and
``annotation_schema_uri``.
:param model_display_name: If the script produces a managed Vertex AI Model. The display name of
the Model. The name can be up to 128 characters long and can be consist
of any UTF-8 characters.
If not provided upon creation, the job's display_name is used.
:param model_labels: Optional. The labels with user-defined metadata to
organize your Models.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param base_output_dir: GCS output directory of job. If not provided a timestamped directory in the
staging directory will be used.
Vertex AI sets the following environment variables when it runs your training code:
- AIP_MODEL_DIR: a Cloud Storage URI of a directory intended for saving model artifacts,
i.e. <base_output_dir>/model/
- AIP_CHECKPOINT_DIR: a Cloud Storage URI of a directory intended for saving checkpoints,
i.e. <base_output_dir>/checkpoints/
- AIP_TENSORBOARD_LOG_DIR: a Cloud Storage URI of a directory intended for saving TensorBoard
logs, i.e. <base_output_dir>/logs/
:param service_account: Specifies the service account for workload run-as account.
Users submitting jobs must have act-as permission on this run-as account.
:param network: The full name of the Compute Engine network to which the job
should be peered.
Private services access must already be configured for the network.
If left unspecified, the job is not peered with any network.
:param bigquery_destination: Provide this field if `dataset` is a BiqQuery dataset.
The BigQuery project location where the training data is to
be written to. In the given project a new dataset is created
with name
``dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>``
where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All
training input data will be written into that dataset. In
the dataset three tables will be created, ``training``,
``validation`` and ``test``.
- AIP_DATA_FORMAT = "bigquery".
- AIP_TRAINING_DATA_URI ="bigquery_destination.dataset_*.training"
- AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_*.validation"
- AIP_TEST_DATA_URI = "bigquery_destination.dataset_*.test"
:param args: Command line arguments to be passed to the Python script.
:param environment_variables: Environment variables to be passed to the container.
Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
At most 10 environment variables can be specified.
The Name of the environment variable must be unique.
:param replica_count: The number of worker replicas. If replica count = 1 then one chief
replica will be provisioned. If replica_count > 1 the remainder will be
provisioned as a worker replica pool.
:param machine_type: The type of machine to use for training.
:param accelerator_type: Hardware accelerator type. One of ACCELERATOR_TYPE_UNSPECIFIED,
NVIDIA_TESLA_K80, NVIDIA_TESLA_P100, NVIDIA_TESLA_V100, NVIDIA_TESLA_P4,
NVIDIA_TESLA_T4
:param accelerator_count: The number of accelerators to attach to a worker replica.
:param 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).
:param boot_disk_size_gb: Size in GB of the boot disk, default is 100GB.
boot disk size must be within the range of [100, 64000].
:param training_fraction_split: Optional. The fraction of the input data that is to be used to train
the Model. This is ignored if Dataset is not provided.
:param validation_fraction_split: Optional. The fraction of the input data that is to be used to
validate the Model. This is ignored if Dataset is not provided.
:param test_fraction_split: Optional. The fraction of the input data that is to be used to evaluate
the Model. This is ignored if Dataset is not provided.
:param training_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to train the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param validation_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to validate the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param test_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to test the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param predefined_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key (either the label's value or
value in the column) must be one of {``training``,
``validation``, ``test``}, and it defines to which set the
given piece of data is assigned. If for a piece of data the
key is not present or has an invalid value, that piece is
ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param timestamp_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key values of the key (the values in
the column) must be in RFC 3339 `date-time` format, where
`time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a
piece of data the key is not present or has an invalid value,
that piece is ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param tensorboard: Optional. The name of a Vertex AI resource to which this CustomJob will upload
logs. Format:
``projects/{project}/locations/{location}/tensorboards/{tensorboard}``
For more information on configuring your service account please visit:
https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-training
"""
self._job = self.get_custom_python_package_training_job(
project=project_id,
location=region,
display_name=display_name,
python_package_gcs_uri=python_package_gcs_uri,
python_module_name=python_module_name,
container_uri=container_uri,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
if not self._job:
raise AirflowException("CustomPythonPackageTrainingJob instance creation failed.")
self._job.run(
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
sync=False,
)
return self._job
@GoogleBaseHook.fallback_to_default_project_id
[docs] def submit_custom_training_job(
self,
*,
project_id: str,
region: str,
display_name: str,
script_path: str,
container_uri: str,
requirements: Sequence[str] | None = None,
model_serving_container_image_uri: str | None = None,
model_serving_container_predict_route: str | None = None,
model_serving_container_health_route: str | None = None,
model_serving_container_command: Sequence[str] | None = None,
model_serving_container_args: Sequence[str] | None = None,
model_serving_container_environment_variables: dict[str, str] | None = None,
model_serving_container_ports: Sequence[int] | None = None,
model_description: str | None = None,
model_instance_schema_uri: str | None = None,
model_parameters_schema_uri: str | None = None,
model_prediction_schema_uri: str | None = None,
parent_model: str | None = None,
is_default_version: bool | None = None,
model_version_aliases: list[str] | None = None,
model_version_description: str | None = None,
labels: dict[str, str] | None = None,
training_encryption_spec_key_name: str | None = None,
model_encryption_spec_key_name: str | None = None,
staging_bucket: str | None = None,
# RUN
dataset: None
| (
datasets.ImageDataset | datasets.TabularDataset | datasets.TextDataset | datasets.VideoDataset
) = None,
annotation_schema_uri: str | None = None,
model_display_name: str | None = None,
model_labels: dict[str, str] | None = None,
base_output_dir: str | None = None,
service_account: str | None = None,
network: str | None = None,
bigquery_destination: str | None = None,
args: list[str | float | int] | None = None,
environment_variables: dict[str, str] | None = None,
replica_count: int = 1,
machine_type: str = "n1-standard-4",
accelerator_type: str = "ACCELERATOR_TYPE_UNSPECIFIED",
accelerator_count: int = 0,
boot_disk_type: str = "pd-ssd",
boot_disk_size_gb: int = 100,
training_fraction_split: float | None = None,
validation_fraction_split: float | None = None,
test_fraction_split: float | None = None,
training_filter_split: str | None = None,
validation_filter_split: str | None = None,
test_filter_split: str | None = None,
predefined_split_column_name: str | None = None,
timestamp_split_column_name: str | None = None,
tensorboard: str | None = None,
) -> CustomTrainingJob:
"""
Create and submit a Custom Training Job pipeline, then exit without waiting for it to complete.
Neither the training model nor backing custom job are available at the moment when the training
pipeline is submitted, both are created only after a period of time. Therefore, it is not possible
to extract and return them in this method, this should be done with a separate client request.
:param display_name: Required. The user-defined name of this TrainingPipeline.
:param script_path: Required. Local path to training script.
:param container_uri: Required: Uri of the training container image in the GCR.
:param requirements: List of python packages dependencies of script.
:param model_serving_container_image_uri: If the training produces a managed Vertex AI Model, the URI
of the Model serving container suitable for serving the model produced by the
training script.
:param model_serving_container_predict_route: If the training produces a managed Vertex AI Model, An
HTTP path to send prediction requests to the container, and which must be supported
by it. If not specified a default HTTP path will be used by Vertex AI.
:param model_serving_container_health_route: If the training produces a managed Vertex AI Model, an
HTTP path to send health check requests to the container, and which must be supported
by it. If not specified a standard HTTP path will be used by AI Platform.
:param model_serving_container_command: The command with which the container is run. Not executed
within a shell. The Docker image's ENTRYPOINT is used if this is not provided.
Variable references $(VAR_NAME) are expanded using the container's
environment. If a variable cannot be resolved, the reference in the
input string will be unchanged. The $(VAR_NAME) syntax can be escaped
with a double $$, ie: $$(VAR_NAME). Escaped references will never be
expanded, regardless of whether the variable exists or not.
:param model_serving_container_args: The arguments to the command. The Docker image's CMD is used if
this is not provided. Variable references $(VAR_NAME) are expanded using the
container's environment. If a variable cannot be resolved, the reference
in the input string will be unchanged. The $(VAR_NAME) syntax can be
escaped with a double $$, ie: $$(VAR_NAME). Escaped references will
never be expanded, regardless of whether the variable exists or not.
:param model_serving_container_environment_variables: The environment variables that are to be
present in the container. Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
:param model_serving_container_ports: Declaration of ports that are exposed by the container. This
field is primarily informational, it gives Vertex AI information about the
network connections the container uses. Listing or not a port here has
no impact on whether the port is actually exposed, any port listening on
the default "0.0.0.0" address inside a container will be accessible from
the network.
:param model_description: The description of the Model.
:param model_instance_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single instance, which
are used in
``PredictRequest.instances``,
``ExplainRequest.instances``
and
``BatchPredictionJob.input_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param model_parameters_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the parameters of prediction and
explanation via
``PredictRequest.parameters``,
``ExplainRequest.parameters``
and
``BatchPredictionJob.model_parameters``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform, if no parameters are supported it is set to an
empty string. Note: The URI given on output will be
immutable and probably different, including the URI scheme,
than the one given on input. The output URI will point to a
location where the user only has a read access.
:param model_prediction_schema_uri: Optional. Points to a YAML file stored on Google Cloud
Storage describing the format of a single prediction
produced by this Model, which are returned via
``PredictResponse.predictions``,
``ExplainResponse.explanations``,
and
``BatchPredictionJob.output_config``.
The schema is defined as an OpenAPI 3.0.2 `Schema
Object <https://tinyurl.com/y538mdwt#schema-object>`__.
AutoML Models always have this field populated by AI
Platform. Note: The URI given on output will be immutable
and probably different, including the URI scheme, than the
one given on input. The output URI will point to a location
where the user only has a read access.
:param parent_model: Optional. The resource name or model ID of an existing model.
The new model uploaded by this job will be a version of `parent_model`.
Only set this field when training a new version of an existing model.
:param is_default_version: Optional. When set to True, the newly uploaded model version will
automatically have alias "default" included. Subsequent uses of
the model produced by this job without a version specified will
use this "default" version.
When set to False, the "default" alias will not be moved.
Actions targeting the model version produced by this job will need
to specifically reference this version by ID or alias.
New model uploads, i.e. version 1, will always be "default" aliased.
:param model_version_aliases: Optional. User provided version aliases so that the model version
uploaded by this job can be referenced via alias instead of
auto-generated version ID. A default version alias will be created
for the first version of the model.
The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9]
:param model_version_description: Optional. The description of the model version
being uploaded by this job.
:param project_id: Project to run training in.
:param region: Location to run training in.
:param labels: Optional. The labels with user-defined metadata to
organize TrainingPipelines.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param training_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the training pipeline. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, this TrainingPipeline will be secured by this key.
Note: Model trained by this TrainingPipeline is also secured
by this key if ``model_to_upload`` is not set separately.
:param model_encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer
managed encryption key used to protect the model. Has the
form:
``projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key``.
The key needs to be in the same region as where the compute
resource is created.
If set, the trained Model will be secured by this key.
:param staging_bucket: Bucket used to stage source and training artifacts.
:param dataset: Vertex AI to fit this training against.
:param annotation_schema_uri: Google Cloud Storage URI points to a YAML file describing
annotation schema. The schema is defined as an OpenAPI 3.0.2
[Schema Object]
(https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schema-object)
Only Annotations that both match this schema and belong to
DataItems not ignored by the split method are used in
respectively training, validation or test role, depending on
the role of the DataItem they are on.
When used in conjunction with
``annotations_filter``,
the Annotations used for training are filtered by both
``annotations_filter``
and
``annotation_schema_uri``.
:param model_display_name: If the script produces a managed Vertex AI Model. The display name of
the Model. The name can be up to 128 characters long and can be consist
of any UTF-8 characters.
If not provided upon creation, the job's display_name is used.
:param model_labels: Optional. The labels with user-defined metadata to
organize your Models.
Label keys and values can be no longer than 64
characters, can only
contain lowercase letters, numeric characters,
underscores and dashes. International characters
are allowed.
See https://goo.gl/xmQnxf for more information
and examples of labels.
:param base_output_dir: GCS output directory of job. If not provided a timestamped directory in the
staging directory will be used.
Vertex AI sets the following environment variables when it runs your training code:
- AIP_MODEL_DIR: a Cloud Storage URI of a directory intended for saving model artifacts,
i.e. <base_output_dir>/model/
- AIP_CHECKPOINT_DIR: a Cloud Storage URI of a directory intended for saving checkpoints,
i.e. <base_output_dir>/checkpoints/
- AIP_TENSORBOARD_LOG_DIR: a Cloud Storage URI of a directory intended for saving TensorBoard
logs, i.e. <base_output_dir>/logs/
:param service_account: Specifies the service account for workload run-as account.
Users submitting jobs must have act-as permission on this run-as account.
:param network: The full name of the Compute Engine network to which the job
should be peered.
Private services access must already be configured for the network.
If left unspecified, the job is not peered with any network.
:param bigquery_destination: Provide this field if `dataset` is a BiqQuery dataset.
The BigQuery project location where the training data is to
be written to. In the given project a new dataset is created
with name
``dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>``
where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All
training input data will be written into that dataset. In
the dataset three tables will be created, ``training``,
``validation`` and ``test``.
- AIP_DATA_FORMAT = "bigquery".
- AIP_TRAINING_DATA_URI ="bigquery_destination.dataset_*.training"
- AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_*.validation"
- AIP_TEST_DATA_URI = "bigquery_destination.dataset_*.test"
:param args: Command line arguments to be passed to the Python script.
:param environment_variables: Environment variables to be passed to the container.
Should be a dictionary where keys are environment variable names
and values are environment variable values for those names.
At most 10 environment variables can be specified.
The Name of the environment variable must be unique.
:param replica_count: The number of worker replicas. If replica count = 1 then one chief
replica will be provisioned. If replica_count > 1 the remainder will be
provisioned as a worker replica pool.
:param machine_type: The type of machine to use for training.
:param accelerator_type: Hardware accelerator type. One of ACCELERATOR_TYPE_UNSPECIFIED,
NVIDIA_TESLA_K80, NVIDIA_TESLA_P100, NVIDIA_TESLA_V100, NVIDIA_TESLA_P4,
NVIDIA_TESLA_T4
:param accelerator_count: The number of accelerators to attach to a worker replica.
:param 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).
:param boot_disk_size_gb: Size in GB of the boot disk, default is 100GB.
boot disk size must be within the range of [100, 64000].
:param training_fraction_split: Optional. The fraction of the input data that is to be used to train
the Model. This is ignored if Dataset is not provided.
:param validation_fraction_split: Optional. The fraction of the input data that is to be used to
validate the Model. This is ignored if Dataset is not provided.
:param test_fraction_split: Optional. The fraction of the input data that is to be used to evaluate
the Model. This is ignored if Dataset is not provided.
:param training_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to train the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param validation_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to validate the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param test_filter_split: Optional. A filter on DataItems of the Dataset. DataItems that match
this filter are used to test the Model. A filter with same syntax
as the one used in DatasetService.ListDataItems may be used. If a
single DataItem is matched by more than one of the FilterSplit filters,
then it is assigned to the first set that applies to it in the training,
validation, test order. This is ignored if Dataset is not provided.
:param predefined_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key (either the label's value or
value in the column) must be one of {``training``,
``validation``, ``test``}, and it defines to which set the
given piece of data is assigned. If for a piece of data the
key is not present or has an invalid value, that piece is
ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param timestamp_split_column_name: Optional. The key is a name of one of the Dataset's data
columns. The value of the key values of the key (the values in
the column) must be in RFC 3339 `date-time` format, where
`time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a
piece of data the key is not present or has an invalid value,
that piece is ignored by the pipeline.
Supported only for tabular and time series Datasets.
:param tensorboard: Optional. The name of a Vertex AI resource to which this CustomJob will upload
logs. Format:
``projects/{project}/locations/{location}/tensorboards/{tensorboard}``
For more information on configuring your service account please visit:
https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-training
"""
self._job = self.get_custom_training_job(
project=project_id,
location=region,
display_name=display_name,
script_path=script_path,
container_uri=container_uri,
requirements=requirements,
model_serving_container_image_uri=model_serving_container_image_uri,
model_serving_container_predict_route=model_serving_container_predict_route,
model_serving_container_health_route=model_serving_container_health_route,
model_serving_container_command=model_serving_container_command,
model_serving_container_args=model_serving_container_args,
model_serving_container_environment_variables=model_serving_container_environment_variables,
model_serving_container_ports=model_serving_container_ports,
model_description=model_description,
model_instance_schema_uri=model_instance_schema_uri,
model_parameters_schema_uri=model_parameters_schema_uri,
model_prediction_schema_uri=model_prediction_schema_uri,
labels=labels,
training_encryption_spec_key_name=training_encryption_spec_key_name,
model_encryption_spec_key_name=model_encryption_spec_key_name,
staging_bucket=staging_bucket,
)
if not self._job:
raise AirflowException("CustomTrainingJob instance creation failed.")
self._job.submit(
dataset=dataset,
annotation_schema_uri=annotation_schema_uri,
model_display_name=model_display_name,
model_labels=model_labels,
base_output_dir=base_output_dir,
service_account=service_account,
network=network,
bigquery_destination=bigquery_destination,
args=args,
environment_variables=environment_variables,
replica_count=replica_count,
machine_type=machine_type,
accelerator_type=accelerator_type,
accelerator_count=accelerator_count,
boot_disk_type=boot_disk_type,
boot_disk_size_gb=boot_disk_size_gb,
training_fraction_split=training_fraction_split,
validation_fraction_split=validation_fraction_split,
test_fraction_split=test_fraction_split,
training_filter_split=training_filter_split,
validation_filter_split=validation_filter_split,
test_filter_split=test_filter_split,
predefined_split_column_name=predefined_split_column_name,
timestamp_split_column_name=timestamp_split_column_name,
tensorboard=tensorboard,
parent_model=parent_model,
is_default_version=is_default_version,
model_version_aliases=model_version_aliases,
model_version_description=model_version_description,
sync=False,
)
return self._job
@GoogleBaseHook.fallback_to_default_project_id
[docs] def delete_training_pipeline(
self,
project_id: str,
region: str,
training_pipeline: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> Operation:
"""
Delete a TrainingPipeline.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param training_pipeline: Required. The name of the TrainingPipeline resource to be deleted.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
name = client.training_pipeline_path(project_id, region, training_pipeline)
result = client.delete_training_pipeline(
request={"name": name},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def delete_custom_job(
self,
project_id: str,
region: str,
custom_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> Operation:
"""
Delete a CustomJob.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param custom_job: Required. The name of the CustomJob to delete.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_job_service_client(region)
name = client.custom_job_path(project_id, region, custom_job)
result = client.delete_custom_job(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
@deprecated(
planned_removal_date="March 01, 2025",
use_instead="PipelineJobHook.get_pipeline_job",
category=AirflowProviderDeprecationWarning,
)
[docs] def get_pipeline_job(
self,
project_id: str,
region: str,
pipeline_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> PipelineJob:
"""
Get a PipelineJob.
This method is deprecated, please use `PipelineJobHook.get_pipeline_job` method.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param pipeline_job: Required. The name of the PipelineJob resource.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
name = client.pipeline_job_path(project_id, region, pipeline_job)
result = client.get_pipeline_job(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def get_training_pipeline(
self,
project_id: str,
region: str,
training_pipeline: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> TrainingPipeline:
"""
Get a TrainingPipeline.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param training_pipeline: Required. The name of the TrainingPipeline resource.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
name = client.training_pipeline_path(project_id, region, training_pipeline)
result = client.get_training_pipeline(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def get_custom_job(
self,
project_id: str,
region: str,
custom_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> CustomJob:
"""
Get a CustomJob.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param custom_job: Required. The name of the CustomJob to get.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_job_service_client(region)
name = JobServiceClient.custom_job_path(project_id, region, custom_job)
result = client.get_custom_job(
request={
"name": name,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
@deprecated(
planned_removal_date="March 01, 2025",
use_instead="PipelineJobHook.list_pipeline_jobs",
category=AirflowProviderDeprecationWarning,
)
[docs] def list_pipeline_jobs(
self,
project_id: str,
region: str,
page_size: int | None = None,
page_token: str | None = None,
filter: str | None = None,
order_by: str | None = None,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> ListPipelineJobsPager:
"""
List PipelineJobs in a Location.
This method is deprecated, please use `PipelineJobHook.list_pipeline_jobs` method.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param filter: Optional. Lists the PipelineJobs that match the filter expression. The
following fields are supported:
- ``pipeline_name``: Supports ``=`` and ``!=`` comparisons.
- ``display_name``: Supports ``=``, ``!=`` comparisons, and
``:`` wildcard.
- ``pipeline_job_user_id``: Supports ``=``, ``!=``
comparisons, and ``:`` wildcard. for example, can check
if pipeline's display_name contains *step* by doing
display_name:"*step*"
- ``create_time``: Supports ``=``, ``!=``, ``<``, ``>``,
``<=``, and ``>=`` comparisons. Values must be in RFC
3339 format.
- ``update_time``: Supports ``=``, ``!=``, ``<``, ``>``,
``<=``, and ``>=`` comparisons. Values must be in RFC
3339 format.
- ``end_time``: Supports ``=``, ``!=``, ``<``, ``>``,
``<=``, and ``>=`` comparisons. Values must be in RFC
3339 format.
- ``labels``: Supports key-value equality and key presence.
Filter expressions can be combined together using logical
operators (``AND`` & ``OR``). For example:
``pipeline_name="test" AND create_time>"2020-05-18T13:30:00Z"``.
The syntax to define filter expression is based on
https://google.aip.dev/160.
:param page_size: Optional. The standard list page size.
:param page_token: Optional. The standard list page token. Typically obtained via
[ListPipelineJobsResponse.next_page_token][google.cloud.aiplatform.v1.ListPipelineJobsResponse.next_page_token]
of the previous
[PipelineService.ListPipelineJobs][google.cloud.aiplatform.v1.PipelineService.ListPipelineJobs]
call.
:param order_by: Optional. A comma-separated list of fields to order by. The default
sort order is in ascending order. Use "desc" after a field
name for descending. You can have multiple order_by fields
provided e.g. "create_time desc, end_time", "end_time,
start_time, update_time" For example, using "create_time
desc, end_time" will order results by create time in
descending order, and if there are multiple jobs having the
same create time, order them by the end time in ascending
order. if order_by is not specified, it will order by
default order is create time in descending order. Supported
fields:
- ``create_time``
- ``update_time``
- ``end_time``
- ``start_time``
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
parent = client.common_location_path(project_id, region)
result = client.list_pipeline_jobs(
request={
"parent": parent,
"page_size": page_size,
"page_token": page_token,
"filter": filter,
"order_by": order_by,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def list_training_pipelines(
self,
project_id: str,
region: str,
page_size: int | None = None,
page_token: str | None = None,
filter: str | None = None,
read_mask: str | None = None,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> ListTrainingPipelinesPager:
"""
List TrainingPipelines in a Location.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param filter: Optional. The standard list filter. Supported fields:
- ``display_name`` supports = and !=.
- ``state`` supports = and !=.
Some examples of using the filter are:
- ``state="PIPELINE_STATE_SUCCEEDED" AND display_name="my_pipeline"``
- ``state="PIPELINE_STATE_RUNNING" OR display_name="my_pipeline"``
- ``NOT display_name="my_pipeline"``
- ``state="PIPELINE_STATE_FAILED"``
:param page_size: Optional. The standard list page size.
:param page_token: Optional. The standard list page token. Typically obtained via
[ListTrainingPipelinesResponse.next_page_token][google.cloud.aiplatform.v1.ListTrainingPipelinesResponse.next_page_token]
of the previous
[PipelineService.ListTrainingPipelines][google.cloud.aiplatform.v1.PipelineService.ListTrainingPipelines]
call.
:param read_mask: Optional. Mask specifying which fields to read.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
parent = client.common_location_path(project_id, region)
result = client.list_training_pipelines(
request={
"parent": parent,
"page_size": page_size,
"page_token": page_token,
"filter": filter,
"read_mask": read_mask,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
[docs] def list_custom_jobs(
self,
project_id: str,
region: str,
page_size: int | None,
page_token: str | None,
filter: str | None,
read_mask: str | None,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> ListCustomJobsPager:
"""
List CustomJobs in a Location.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param filter: Optional. The standard list filter. Supported fields:
- ``display_name`` supports = and !=.
- ``state`` supports = and !=.
Some examples of using the filter are:
- ``state="PIPELINE_STATE_SUCCEEDED" AND display_name="my_pipeline"``
- ``state="PIPELINE_STATE_RUNNING" OR display_name="my_pipeline"``
- ``NOT display_name="my_pipeline"``
- ``state="PIPELINE_STATE_FAILED"``
:param page_size: Optional. The standard list page size.
:param page_token: Optional. The standard list page token. Typically obtained via
[ListTrainingPipelinesResponse.next_page_token][google.cloud.aiplatform.v1.ListTrainingPipelinesResponse.next_page_token]
of the previous
[PipelineService.ListTrainingPipelines][google.cloud.aiplatform.v1.PipelineService.ListTrainingPipelines]
call.
:param read_mask: Optional. Mask specifying which fields to read.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_job_service_client(region)
parent = JobServiceClient.common_location_path(project_id, region)
result = client.list_custom_jobs(
request={
"parent": parent,
"page_size": page_size,
"page_token": page_token,
"filter": filter,
"read_mask": read_mask,
},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
@GoogleBaseHook.fallback_to_default_project_id
@deprecated(
planned_removal_date="March 01, 2025",
use_instead="PipelineJobHook.delete_pipeline_job",
category=AirflowProviderDeprecationWarning,
)
[docs] def delete_pipeline_job(
self,
project_id: str,
region: str,
pipeline_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
) -> Operation:
"""
Delete a PipelineJob.
This method is deprecated, please use `PipelineJobHook.delete_pipeline_job` method.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param region: Required. The ID of the Google Cloud region that the service belongs to.
:param pipeline_job: Required. The name of the PipelineJob resource to be deleted.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
"""
client = self.get_pipeline_service_client(region)
name = client.pipeline_job_path(project_id, region, pipeline_job)
result = client.delete_pipeline_job(
request={"name": name},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
[docs]class CustomJobAsyncHook(GoogleBaseAsyncHook):
"""Async hook for Custom Job Service Client."""
[docs] sync_hook_class = CustomJobHook
[docs] JOB_COMPLETE_STATES = {
JobState.JOB_STATE_CANCELLED,
JobState.JOB_STATE_FAILED,
JobState.JOB_STATE_PAUSED,
JobState.JOB_STATE_SUCCEEDED,
}
[docs] PIPELINE_COMPLETE_STATES = (
PipelineState.PIPELINE_STATE_CANCELLED,
PipelineState.PIPELINE_STATE_FAILED,
PipelineState.PIPELINE_STATE_PAUSED,
PipelineState.PIPELINE_STATE_SUCCEEDED,
)
def __init__(
self,
gcp_conn_id: str = "google_cloud_default",
impersonation_chain: str | Sequence[str] | None = None,
**kwargs,
):
super().__init__(
gcp_conn_id=gcp_conn_id,
impersonation_chain=impersonation_chain,
**kwargs,
)
self._job: None | (
CustomContainerTrainingJob | CustomPythonPackageTrainingJob | CustomTrainingJob
) = None
[docs] async def get_credentials(self) -> Credentials:
return (await self.get_sync_hook()).get_credentials()
[docs] async def get_job_service_client(
self,
region: str | None = None,
) -> JobServiceAsyncClient:
"""Retrieve Vertex AI JobServiceAsyncClient object."""
if region and region != "global":
client_options = ClientOptions(api_endpoint=f"{region}-aiplatform.googleapis.com:443")
else:
client_options = ClientOptions()
return JobServiceAsyncClient(
credentials=(await self.get_credentials()),
client_info=CLIENT_INFO,
client_options=client_options,
)
[docs] async def get_pipeline_service_client(
self,
region: str | None = None,
) -> PipelineServiceAsyncClient:
"""Retrieve Vertex AI PipelineServiceAsyncClient object."""
if region and region != "global":
client_options = ClientOptions(api_endpoint=f"{region}-aiplatform.googleapis.com:443")
else:
client_options = ClientOptions()
return PipelineServiceAsyncClient(
credentials=(await self.get_credentials()),
client_info=CLIENT_INFO,
client_options=client_options,
)
[docs] async def get_custom_job(
self,
project_id: str,
location: str,
job_id: str,
retry: AsyncRetry | _MethodDefault = DEFAULT,
timeout: float | _MethodDefault | None = DEFAULT,
metadata: Sequence[tuple[str, str]] = (),
client: JobServiceAsyncClient | None = None,
) -> types.CustomJob:
"""
Get a CustomJob proto message from JobServiceAsyncClient.
:param project_id: Required. The ID of the Google Cloud project that the job belongs to.
:param location: Required. The ID of the Google Cloud region that the job belongs to.
:param job_id: Required. The custom job id.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
:param client: The async job service client.
"""
if not client:
client = await self.get_job_service_client(region=location)
job_name = client.custom_job_path(project_id, location, job_id)
result: types.CustomJob = await client.get_custom_job(
request={"name": job_name},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return result
[docs] async def get_training_pipeline(
self,
project_id: str,
location: str,
pipeline_id: str,
retry: AsyncRetry | _MethodDefault = DEFAULT,
timeout: float | _MethodDefault | None = DEFAULT,
metadata: Sequence[tuple[str, str]] = (),
client: PipelineServiceAsyncClient | None = None,
) -> types.TrainingPipeline:
"""
Get a TrainingPipeline proto message from PipelineServiceAsyncClient.
:param project_id: Required. The ID of the Google Cloud project that the service belongs to.
:param location: Required. The ID of the Google Cloud region that the service belongs to.
:param pipeline_id: Required. The ID of the PipelineJob resource.
:param retry: Designation of what errors, if any, should be retried.
:param timeout: The timeout for this request.
:param metadata: Strings which should be sent along with the request as metadata.
:param client: The async pipeline service client.
"""
if not client:
client = await self.get_pipeline_service_client(region=location)
pipeline_name = client.training_pipeline_path(
project=project_id,
location=location,
training_pipeline=pipeline_id,
)
response: types.TrainingPipeline = await client.get_training_pipeline(
request={"name": pipeline_name},
retry=retry,
timeout=timeout,
metadata=metadata,
)
return response
[docs] async def wait_for_custom_job(
self,
project_id: str,
location: str,
job_id: str,
retry: AsyncRetry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
poll_interval: int = 10,
) -> types.CustomJob:
"""Make async calls to Vertex AI to check the custom job state until it is complete."""
client = await self.get_job_service_client(region=location)
while True:
try:
self.log.info("Requesting a custom job with id %s", job_id)
job: types.CustomJob = await self.get_custom_job(
project_id=project_id,
location=location,
job_id=job_id,
retry=retry,
timeout=timeout,
metadata=metadata,
client=client,
)
except Exception as ex:
self.log.exception("Exception occurred while requesting job %s", job_id)
raise AirflowException(ex)
self.log.info("Status of the custom job %s is %s", job.name, job.state.name)
if job.state in self.JOB_COMPLETE_STATES:
return job
self.log.info("Sleeping for %s seconds.", poll_interval)
await asyncio.sleep(poll_interval)
[docs] async def wait_for_training_pipeline(
self,
project_id: str,
location: str,
pipeline_id: str,
retry: AsyncRetry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
poll_interval: int = 10,
) -> types.TrainingPipeline:
"""Make async calls to Vertex AI to check the training pipeline state until it is complete."""
client = await self.get_pipeline_service_client(region=location)
while True:
try:
self.log.info("Requesting a training pipeline with id %s", pipeline_id)
pipeline: types.TrainingPipeline = await self.get_training_pipeline(
project_id=project_id,
location=location,
pipeline_id=pipeline_id,
retry=retry,
timeout=timeout,
metadata=metadata,
client=client,
)
except Exception as ex:
self.log.exception("Exception occurred while requesting training pipeline %s", pipeline_id)
raise AirflowException(ex)
self.log.info("Status of the training pipeline %s is %s", pipeline.name, pipeline.state.name)
if pipeline.state in self.PIPELINE_COMPLETE_STATES:
return pipeline
self.log.info("Sleeping for %s seconds.", poll_interval)
await asyncio.sleep(poll_interval)