#
# 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 Google Vertex AI operators."""
from __future__ import annotations
import warnings
from functools import cached_property
from typing import TYPE_CHECKING, Any, Sequence
from google.api_core.exceptions import NotFound
from google.api_core.gapic_v1.method import DEFAULT, _MethodDefault
from google.cloud.aiplatform_v1.types import BatchPredictionJob
from airflow.configuration import conf
from airflow.exceptions import AirflowException, AirflowProviderDeprecationWarning
from airflow.providers.google.cloud.hooks.vertex_ai.batch_prediction_job import BatchPredictionJobHook
from airflow.providers.google.cloud.links.vertex_ai import (
VertexAIBatchPredictionJobLink,
VertexAIBatchPredictionJobListLink,
)
from airflow.providers.google.cloud.operators.cloud_base import GoogleCloudBaseOperator
from airflow.providers.google.cloud.triggers.vertex_ai import CreateBatchPredictionJobTrigger
if TYPE_CHECKING:
from google.api_core.retry import Retry
from google.cloud.aiplatform import BatchPredictionJob as BatchPredictionJobObject, Model, explain
from airflow.utils.context import Context
[docs]class CreateBatchPredictionJobOperator(GoogleCloudBaseOperator):
"""
Creates a BatchPredictionJob. A BatchPredictionJob once created will right away be attempted to start.
: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 batch_prediction_job: Required. The BatchPredictionJob to create.
:param job_display_name: Required. The user-defined name of the BatchPredictionJob. The name can be
up to 128 characters long and can consist of any UTF-8 characters.
:param model_name: Required. A fully-qualified model resource name or model ID.
:param instances_format: Required. The format in which instances are provided. Must be one of the
formats listed in `Model.supported_input_storage_formats`. Default is "jsonl" when using
`gcs_source`. If a `bigquery_source` is provided, this is overridden to "bigquery".
:param predictions_format: Required. The format in which Vertex AI outputs the predictions, must be
one of the formats specified in `Model.supported_output_storage_formats`. Default is "jsonl" when
using `gcs_destination_prefix`. If a `bigquery_destination_prefix` is provided, this is
overridden to "bigquery".
:param gcs_source: Google Cloud Storage URI(-s) to your instances to run batch prediction on. They
must match `instances_format`. May contain wildcards. For more information on wildcards, see
https://cloud.google.com/storage/docs/gsutil/addlhelp/WildcardNames.
:param bigquery_source: BigQuery URI to a table, up to 2000 characters long.
For example: `bq://projectId.bqDatasetId.bqTableId`
:param gcs_destination_prefix: The Google Cloud Storage location of the directory where the output is
to be written to. In the given directory a new directory is created. Its name is
``prediction-<model-display-name>-<job-create-time>``, where timestamp is in
YYYY-MM-DDThh:mm:ss.sssZ ISO-8601 format. Inside of it files ``predictions_0001.<extension>``,
``predictions_0002.<extension>``, ..., ``predictions_N.<extension>`` are created where
``<extension>`` depends on chosen ``predictions_format``, and N may equal 0001 and depends on the
total number of successfully predicted instances. If the Model has both ``instance`` and
``prediction`` schemata defined then each such file contains predictions as per the
``predictions_format``. If prediction for any instance failed (partially or completely), then an
additional ``errors_0001.<extension>``, ``errors_0002.<extension>``,..., ``errors_N.<extension>``
files are created (N depends on total number of failed predictions). These files contain the
failed instances, as per their schema, followed by an additional ``error`` field which as value
has ```google.rpc.Status`` <Status>`__ containing only ``code`` and ``message`` fields.
:param bigquery_destination_prefix: The BigQuery project location where the output is to be written
to. In the given project a new dataset is created with name
``prediction_<model-display-name>_<job-create-time>`` where is made BigQuery-dataset-name
compatible (for example, most special characters become underscores), and timestamp is in
YYYY_MM_DDThh_mm_ss_sssZ "based on ISO-8601" format. In the dataset two tables will be created,
``predictions``, and ``errors``. If the Model has both ``instance`` and ``prediction`` schemata
defined then the tables have columns as follows: The ``predictions`` table contains instances for
which the prediction succeeded, it has columns as per a concatenation of the Model's instance and
prediction schemata. The ``errors`` table contains rows for which the prediction has failed, it
has instance columns, as per the instance schema, followed by a single "errors" column, which as
values has ```google.rpc.Status`` <Status>`__ represented as a STRUCT, and containing only
``code`` and ``message``.
:param model_parameters: The parameters that govern the predictions. The schema of the parameters may
be specified via the Model's `parameters_schema_uri`.
:param machine_type: The type of machine for running batch prediction on dedicated resources. Not
specifying machine type will result in batch prediction job being run with automatic resources.
:param accelerator_type: The type of accelerator(s) that may be attached to the machine as per
`accelerator_count`. Only used if `machine_type` is set.
:param accelerator_count: The number of accelerators to attach to the `machine_type`. Only used if
`machine_type` is set.
:param starting_replica_count: The number of machine replicas used at the start of the batch
operation. If not set, Vertex AI decides starting number, not greater than `max_replica_count`.
Only used if `machine_type` is set.
:param max_replica_count: The maximum number of machine replicas the batch operation may be scaled
to. Only used if `machine_type` is set. Default is 10.
:param generate_explanation: Optional. Generate explanation along with the batch prediction results.
This will cause the batch prediction output to include explanations based on the
`prediction_format`:
- `bigquery`: output includes a column named `explanation`. The value is a struct that conforms to
the [aiplatform.gapic.Explanation] object.
- `jsonl`: The JSON objects on each line include an additional entry keyed `explanation`. The value
of the entry is a JSON object that conforms to the [aiplatform.gapic.Explanation] object.
- `csv`: Generating explanations for CSV format is not supported.
:param explanation_metadata: Optional. Explanation metadata configuration for this
BatchPredictionJob. Can be specified only if `generate_explanation` is set to `True`.
This value overrides the value of `Model.explanation_metadata`. All fields of
`explanation_metadata` are optional in the request. If a field of the `explanation_metadata`
object is not populated, the corresponding field of the `Model.explanation_metadata` object is
inherited. For more details, see `Ref docs <http://tinyurl.com/1igh60kt>`
:param explanation_parameters: Optional. Parameters to configure explaining for Model's predictions.
Can be specified only if `generate_explanation` is set to `True`.
This value overrides the value of `Model.explanation_parameters`. All fields of
`explanation_parameters` are optional in the request. If a field of the `explanation_parameters`
object is not populated, the corresponding field of the `Model.explanation_parameters` object is
inherited. For more details, see `Ref docs <http://tinyurl.com/1an4zake>`
:param labels: Optional. The labels with user-defined metadata to organize your BatchPredictionJobs.
Label keys and values can be no longer than 64 characters (Unicode codepoints), 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 encryption_spec_key_name: Optional. The Cloud KMS resource identifier of the customer managed
encryption key used to protect the job. 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 this is set, then all resources created by the BatchPredictionJob will be encrypted with the
provided encryption key.
Overrides encryption_spec_key_name set in aiplatform.init.
:param sync: (Deprecated) Whether to execute this method 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.
:param create_request_timeout: Optional. The timeout for the create request in seconds.
:param batch_size: Optional. The number of the records (e.g. instances)
of the operation given in each batch
to a machine replica. Machine type, and size of a single record should be considered
when setting this parameter, higher value speeds up the batch operation's execution,
but too high value will result in a whole batch not fitting in a machine's memory,
and the whole operation will fail.
The default value is same as in the aiplatform's BatchPredictionJob.
: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 gcp_conn_id: The connection ID to use connecting to Google Cloud.
:param impersonation_chain: Optional service account to impersonate using short-term
credentials, or chained list of accounts required to get the access_token
of the last account in the list, which will be impersonated in the request.
If set as a string, the account must grant the originating account
the Service Account Token Creator IAM role.
If set as a sequence, the identities from the list must grant
Service Account Token Creator IAM role to the directly preceding identity, with first
account from the list granting this role to the originating account (templated).
:param deferrable: Optional. Run operator in the deferrable mode.
:param poll_interval: Interval size which defines how often job status is checked in deferrable mode.
"""
[docs] template_fields = ("region", "project_id", "model_name", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
job_display_name: str,
model_name: str | Model,
instances_format: str = "jsonl",
predictions_format: str = "jsonl",
gcs_source: str | Sequence[str] | None = None,
bigquery_source: str | None = None,
gcs_destination_prefix: str | None = None,
bigquery_destination_prefix: str | None = None,
model_parameters: dict | None = None,
machine_type: str | None = None,
accelerator_type: str | None = None,
accelerator_count: int | None = None,
starting_replica_count: int | None = None,
max_replica_count: int | None = None,
generate_explanation: bool | None = False,
explanation_metadata: explain.ExplanationMetadata | None = None,
explanation_parameters: explain.ExplanationParameters | None = None,
labels: dict[str, str] | None = None,
encryption_spec_key_name: str | None = None,
sync: bool = True,
create_request_timeout: float | None = None,
batch_size: int | None = None,
gcp_conn_id: str = "google_cloud_default",
impersonation_chain: str | Sequence[str] | None = None,
deferrable: bool = conf.getboolean("operators", "default_deferrable", fallback=False),
poll_interval: int = 10,
**kwargs,
) -> None:
super().__init__(**kwargs)
self.region = region
self.project_id = project_id
self.job_display_name = job_display_name
self.model_name = model_name
self.instances_format = instances_format
self.predictions_format = predictions_format
self.gcs_source = gcs_source
self.bigquery_source = bigquery_source
self.gcs_destination_prefix = gcs_destination_prefix
self.bigquery_destination_prefix = bigquery_destination_prefix
self.model_parameters = model_parameters
self.machine_type = machine_type
self.accelerator_type = accelerator_type
self.accelerator_count = accelerator_count
self.starting_replica_count = starting_replica_count
self.max_replica_count = max_replica_count
self.generate_explanation = generate_explanation
self.explanation_metadata = explanation_metadata
self.explanation_parameters = explanation_parameters
self.labels = labels
self.encryption_spec_key_name = encryption_spec_key_name
self.sync = sync
self.create_request_timeout = create_request_timeout
self.batch_size = batch_size
self.gcp_conn_id = gcp_conn_id
self.impersonation_chain = impersonation_chain
self.deferrable = deferrable
self.poll_interval = poll_interval
@cached_property
[docs] def hook(self) -> BatchPredictionJobHook:
return BatchPredictionJobHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
[docs] def execute(self, context: Context):
warnings.warn(
"The 'sync' parameter is deprecated and will be removed after 28.08.2024.",
AirflowProviderDeprecationWarning,
stacklevel=2,
)
self.log.info("Creating Batch prediction job")
batch_prediction_job: BatchPredictionJobObject = self.hook.submit_batch_prediction_job(
region=self.region,
project_id=self.project_id,
job_display_name=self.job_display_name,
model_name=self.model_name,
instances_format=self.instances_format,
predictions_format=self.predictions_format,
gcs_source=self.gcs_source,
bigquery_source=self.bigquery_source,
gcs_destination_prefix=self.gcs_destination_prefix,
bigquery_destination_prefix=self.bigquery_destination_prefix,
model_parameters=self.model_parameters,
machine_type=self.machine_type,
accelerator_type=self.accelerator_type,
accelerator_count=self.accelerator_count,
starting_replica_count=self.starting_replica_count,
max_replica_count=self.max_replica_count,
generate_explanation=self.generate_explanation,
explanation_metadata=self.explanation_metadata,
explanation_parameters=self.explanation_parameters,
labels=self.labels,
encryption_spec_key_name=self.encryption_spec_key_name,
create_request_timeout=self.create_request_timeout,
batch_size=self.batch_size,
)
batch_prediction_job.wait_for_resource_creation()
batch_prediction_job_id = batch_prediction_job.name
self.log.info("Batch prediction job was created. Job id: %s", batch_prediction_job_id)
self.xcom_push(context, key="batch_prediction_job_id", value=batch_prediction_job_id)
VertexAIBatchPredictionJobLink.persist(
context=context, task_instance=self, batch_prediction_job_id=batch_prediction_job_id
)
if self.deferrable:
self.defer(
trigger=CreateBatchPredictionJobTrigger(
conn_id=self.gcp_conn_id,
project_id=self.project_id,
location=self.region,
job_id=batch_prediction_job.name,
poll_interval=self.poll_interval,
impersonation_chain=self.impersonation_chain,
),
method_name="execute_complete",
)
batch_prediction_job.wait_for_completion()
self.log.info("Batch prediction job was completed. Job id: %s", batch_prediction_job_id)
return batch_prediction_job.to_dict()
[docs] def on_kill(self) -> None:
"""Act as a callback called when the operator is killed; cancel any running job."""
if self.hook:
self.hook.cancel_batch_prediction_job()
[docs] def execute_complete(self, context: Context, event: dict[str, Any]) -> dict[str, Any]:
if event and event["status"] == "error":
raise AirflowException(event["message"])
job: dict[str, Any] = event["job"]
self.log.info("Batch prediction job %s created and completed successfully.", job["name"])
job_id = self.hook.extract_batch_prediction_job_id(job)
self.xcom_push(
context,
key="batch_prediction_job_id",
value=job_id,
)
self.xcom_push(
context,
key="training_conf",
value={
"training_conf_id": job_id,
"region": self.region,
"project_id": self.project_id,
},
)
return event["job"]
[docs]class DeleteBatchPredictionJobOperator(GoogleCloudBaseOperator):
"""
Deletes a BatchPredictionJob. Can only be called on jobs that already finished.
: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 batch_prediction_job_id: The ID of the BatchPredictionJob 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.
:param gcp_conn_id: The connection ID to use connecting to Google Cloud.
:param impersonation_chain: Optional service account to impersonate using short-term
credentials, or chained list of accounts required to get the access_token
of the last account in the list, which will be impersonated in the request.
If set as a string, the account must grant the originating account
the Service Account Token Creator IAM role.
If set as a sequence, the identities from the list must grant
Service Account Token Creator IAM role to the directly preceding identity, with first
account from the list granting this role to the originating account (templated).
"""
[docs] template_fields = ("region", "project_id", "batch_prediction_job_id", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
batch_prediction_job_id: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
gcp_conn_id: str = "google_cloud_default",
impersonation_chain: str | Sequence[str] | None = None,
**kwargs,
) -> None:
super().__init__(**kwargs)
self.region = region
self.project_id = project_id
self.batch_prediction_job_id = batch_prediction_job_id
self.retry = retry
self.timeout = timeout
self.metadata = metadata
self.gcp_conn_id = gcp_conn_id
self.impersonation_chain = impersonation_chain
[docs] def execute(self, context: Context):
hook = BatchPredictionJobHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
try:
self.log.info("Deleting batch prediction job: %s", self.batch_prediction_job_id)
operation = hook.delete_batch_prediction_job(
project_id=self.project_id,
region=self.region,
batch_prediction_job=self.batch_prediction_job_id,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
hook.wait_for_operation(timeout=self.timeout, operation=operation)
self.log.info("Batch prediction job was deleted.")
except NotFound:
self.log.info("The Batch prediction job %s does not exist.", self.batch_prediction_job_id)
[docs]class GetBatchPredictionJobOperator(GoogleCloudBaseOperator):
"""
Gets a BatchPredictionJob.
: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 batch_prediction_job: Required. The name of the BatchPredictionJob 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 gcp_conn_id: The connection ID to use connecting to Google Cloud.
:param impersonation_chain: Optional service account to impersonate using short-term
credentials, or chained list of accounts required to get the access_token
of the last account in the list, which will be impersonated in the request.
If set as a string, the account must grant the originating account
the Service Account Token Creator IAM role.
If set as a sequence, the identities from the list must grant
Service Account Token Creator IAM role to the directly preceding identity, with first
account from the list granting this role to the originating account (templated).
"""
[docs] template_fields = ("region", "project_id", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
batch_prediction_job: str,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
gcp_conn_id: str = "google_cloud_default",
impersonation_chain: str | Sequence[str] | None = None,
**kwargs,
) -> None:
super().__init__(**kwargs)
self.region = region
self.project_id = project_id
self.batch_prediction_job = batch_prediction_job
self.retry = retry
self.timeout = timeout
self.metadata = metadata
self.gcp_conn_id = gcp_conn_id
self.impersonation_chain = impersonation_chain
[docs] def execute(self, context: Context):
hook = BatchPredictionJobHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
try:
self.log.info("Get batch prediction job: %s", self.batch_prediction_job)
result = hook.get_batch_prediction_job(
project_id=self.project_id,
region=self.region,
batch_prediction_job=self.batch_prediction_job,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
self.log.info("Batch prediction job was gotten.")
VertexAIBatchPredictionJobLink.persist(
context=context, task_instance=self, batch_prediction_job_id=self.batch_prediction_job
)
return BatchPredictionJob.to_dict(result)
except NotFound:
self.log.info("The Batch prediction job %s does not exist.", self.batch_prediction_job)
[docs]class ListBatchPredictionJobsOperator(GoogleCloudBaseOperator):
"""
Lists BatchPredictionJobs 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: The standard list filter.
Supported fields:
- ``display_name`` supports = and !=.
- ``state`` supports = and !=.
- ``model_display_name`` supports = and !=
Some examples of using the filter are:
- ``state="JOB_STATE_SUCCEEDED" AND display_name="my_job"``
- ``state="JOB_STATE_RUNNING" OR display_name="my_job"``
- ``NOT display_name="my_job"``
- ``state="JOB_STATE_FAILED"``
:param page_size: The standard list page size.
:param page_token: The standard list page token.
:param read_mask: 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.
:param gcp_conn_id: The connection ID to use connecting to Google Cloud.
:param impersonation_chain: Optional service account to impersonate using short-term
credentials, or chained list of accounts required to get the access_token
of the last account in the list, which will be impersonated in the request.
If set as a string, the account must grant the originating account
the Service Account Token Creator IAM role.
If set as a sequence, the identities from the list must grant
Service Account Token Creator IAM role to the directly preceding identity, with first
account from the list granting this role to the originating account (templated).
"""
[docs] template_fields = ("region", "project_id", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
filter: str | None = None,
page_size: int | None = None,
page_token: str | None = None,
read_mask: str | None = None,
retry: Retry | _MethodDefault = DEFAULT,
timeout: float | None = None,
metadata: Sequence[tuple[str, str]] = (),
gcp_conn_id: str = "google_cloud_default",
impersonation_chain: str | Sequence[str] | None = None,
**kwargs,
) -> None:
super().__init__(**kwargs)
self.region = region
self.project_id = project_id
self.filter = filter
self.page_size = page_size
self.page_token = page_token
self.read_mask = read_mask
self.retry = retry
self.timeout = timeout
self.metadata = metadata
self.gcp_conn_id = gcp_conn_id
self.impersonation_chain = impersonation_chain
[docs] def execute(self, context: Context):
hook = BatchPredictionJobHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
results = hook.list_batch_prediction_jobs(
project_id=self.project_id,
region=self.region,
filter=self.filter,
page_size=self.page_size,
page_token=self.page_token,
read_mask=self.read_mask,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
VertexAIBatchPredictionJobListLink.persist(context=context, task_instance=self)
return [BatchPredictionJob.to_dict(result) for result in results]