#
# 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
from typing import TYPE_CHECKING, 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 DeployedModel, Endpoint, endpoint_service
from airflow.providers.google.cloud.hooks.vertex_ai.endpoint_service import EndpointServiceHook
from airflow.providers.google.cloud.links.vertex_ai import (
VertexAIEndpointLink,
VertexAIEndpointListLink,
VertexAIModelLink,
)
from airflow.providers.google.cloud.operators.cloud_base import GoogleCloudBaseOperator
if TYPE_CHECKING:
from google.api_core.retry import Retry
from google.protobuf.field_mask_pb2 import FieldMask
from airflow.utils.context import Context
[docs]class CreateEndpointOperator(GoogleCloudBaseOperator):
"""
Creates an Endpoint.
: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 endpoint: Required. The Endpoint to create.
:param endpoint_id: The ID of Endpoint. This value should be 1-10 characters, and valid characters
are /[0-9]/. If not provided, Vertex AI will generate a value for this 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 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,
endpoint: Endpoint | dict,
endpoint_id: 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.endpoint = endpoint
self.endpoint_id = endpoint_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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
self.log.info("Creating endpoint")
operation = hook.create_endpoint(
project_id=self.project_id,
region=self.region,
endpoint=self.endpoint,
endpoint_id=self.endpoint_id,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
result = hook.wait_for_operation(timeout=self.timeout, operation=operation)
endpoint = Endpoint.to_dict(result)
endpoint_id = hook.extract_endpoint_id(endpoint)
self.log.info("Endpoint was created. Endpoint ID: %s", endpoint_id)
self.xcom_push(context, key="endpoint_id", value=endpoint_id)
VertexAIEndpointLink.persist(context=context, task_instance=self, endpoint_id=endpoint_id)
return endpoint
[docs]class DeleteEndpointOperator(GoogleCloudBaseOperator):
"""
Deletes an Endpoint.
: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 endpoint_id: Required. The Endpoint ID 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.
: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", "endpoint_id", "project_id", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
endpoint_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.endpoint_id = endpoint_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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
try:
self.log.info("Deleting endpoint: %s", self.endpoint_id)
operation = hook.delete_endpoint(
project_id=self.project_id,
region=self.region,
endpoint=self.endpoint_id,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
hook.wait_for_operation(timeout=self.timeout, operation=operation)
self.log.info("Endpoint was deleted.")
except NotFound:
self.log.info("The Endpoint ID %s does not exist.", self.endpoint_id)
[docs]class DeployModelOperator(GoogleCloudBaseOperator):
"""
Deploys a Model into this Endpoint, creating a DeployedModel within it.
: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 endpoint_id: Required. The name of the Endpoint resource into which to deploy a Model. Format:
``projects/{project}/locations/{location}/endpoints/{endpoint}``
:param deployed_model: Required. The DeployedModel to be created within the Endpoint. Note that
[Endpoint.traffic_split][google.cloud.aiplatform.v1.Endpoint.traffic_split] must be updated for
the DeployedModel to start receiving traffic, either as part of this call, or via
[EndpointService.UpdateEndpoint][google.cloud.aiplatform.v1.EndpointService.UpdateEndpoint].
:param traffic_split: A map from a DeployedModel's ID to the percentage of this Endpoint's traffic
that should be forwarded to that DeployedModel.
If this field is non-empty, then the Endpoint's
[traffic_split][google.cloud.aiplatform.v1.Endpoint.traffic_split] will be overwritten with it. To
refer to the ID of the just being deployed Model, a "0" should be used, and the actual ID of the
new DeployedModel will be filled in its place by this method. The traffic percentage values must
add up to 100.
If this field is empty, then the Endpoint's
[traffic_split][google.cloud.aiplatform.v1.Endpoint.traffic_split] is not updated.
: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", "endpoint_id", "project_id", "deployed_model", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
endpoint_id: str,
deployed_model: DeployedModel | dict,
traffic_split: Sequence | dict | 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.endpoint_id = endpoint_id
self.deployed_model = deployed_model
self.traffic_split = traffic_split
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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
self.log.info("Deploying model")
operation = hook.deploy_model(
project_id=self.project_id,
region=self.region,
endpoint=self.endpoint_id,
deployed_model=self.deployed_model,
traffic_split=self.traffic_split,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
result = hook.wait_for_operation(timeout=self.timeout, operation=operation)
deploy_model = endpoint_service.DeployModelResponse.to_dict(result)
deployed_model_id = hook.extract_deployed_model_id(deploy_model)
self.log.info("Model was deployed. Deployed Model ID: %s", deployed_model_id)
self.xcom_push(context, key="deployed_model_id", value=deployed_model_id)
VertexAIModelLink.persist(context=context, task_instance=self, model_id=deployed_model_id)
return deploy_model
[docs]class GetEndpointOperator(GoogleCloudBaseOperator):
"""
Gets an Endpoint.
: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 endpoint_id: Required. The Endpoint ID 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.
: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", "endpoint_id", "project_id", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
endpoint_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.endpoint_id = endpoint_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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
try:
self.log.info("Get endpoint: %s", self.endpoint_id)
endpoint_obj = hook.get_endpoint(
project_id=self.project_id,
region=self.region,
endpoint=self.endpoint_id,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
VertexAIEndpointLink.persist(context=context, task_instance=self, endpoint_id=self.endpoint_id)
self.log.info("Endpoint was gotten.")
return Endpoint.to_dict(endpoint_obj)
except NotFound:
self.log.info("The Endpoint ID %s does not exist.", self.endpoint_id)
[docs]class ListEndpointsOperator(GoogleCloudBaseOperator):
"""
Lists Endpoints 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 order_by: A comma-separated list of fields to order by, sorted in
ascending order. Use "desc" after a field name for
descending. Supported fields:
- ``display_name``
- ``create_time``
- ``update_time``
Example: ``display_name, create_time desc``.
: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,
order_by: 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.order_by = order_by
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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
results = hook.list_endpoints(
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,
order_by=self.order_by,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
VertexAIEndpointListLink.persist(context=context, task_instance=self)
return [Endpoint.to_dict(result) for result in results]
[docs]class UndeployModelOperator(GoogleCloudBaseOperator):
"""
Undeploys a Model from an Endpoint, removing a DeployedModel from it, and freeing all used resources.
: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 endpoint_id: Required. The ID of the Endpoint resource from which to undeploy a Model.
:param deployed_model_id: Required. The ID of the DeployedModel to be undeployed from the Endpoint.
:param traffic_split: If this field is provided, then the Endpoint's
[traffic_split][google.cloud.aiplatform.v1.Endpoint.traffic_split] will be overwritten with it. If
last DeployedModel is being undeployed from the Endpoint, the [Endpoint.traffic_split] will always
end up empty when this call returns. A DeployedModel will be successfully undeployed only if it
doesn't have any traffic assigned to it when this method executes, or if this field unassigns any
traffic to it.
: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", "endpoint_id", "deployed_model_id", "project_id", "impersonation_chain")
def __init__(
self,
*,
region: str,
project_id: str,
endpoint_id: str,
deployed_model_id: str,
traffic_split: Sequence | dict | 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.endpoint_id = endpoint_id
self.deployed_model_id = deployed_model_id
self.traffic_split = traffic_split
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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
self.log.info("Removing a DeployedModel %s", self.deployed_model_id)
operation = hook.undeploy_model(
project_id=self.project_id,
region=self.region,
endpoint=self.endpoint_id,
deployed_model_id=self.deployed_model_id,
traffic_split=self.traffic_split,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
hook.wait_for_operation(timeout=self.timeout, operation=operation)
self.log.info("DeployedModel was removed successfully")
[docs]class UpdateEndpointOperator(GoogleCloudBaseOperator):
"""
Updates an Endpoint.
: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 endpoint_id: Required. The ID of the Endpoint to update.
:param endpoint: Required. The Endpoint which replaces the resource on the server.
:param update_mask: Required. The update mask applies to the resource. See
[google.protobuf.FieldMask][google.protobuf.FieldMask].
: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", "endpoint_id", "project_id", "impersonation_chain")
def __init__(
self,
*,
project_id: str,
region: str,
endpoint_id: str,
endpoint: Endpoint | dict,
update_mask: FieldMask | dict,
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.project_id = project_id
self.region = region
self.endpoint_id = endpoint_id
self.endpoint = endpoint
self.update_mask = update_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 = EndpointServiceHook(
gcp_conn_id=self.gcp_conn_id,
impersonation_chain=self.impersonation_chain,
)
self.log.info("Updating endpoint: %s", self.endpoint_id)
result = hook.update_endpoint(
project_id=self.project_id,
region=self.region,
endpoint_id=self.endpoint_id,
endpoint=self.endpoint,
update_mask=self.update_mask,
retry=self.retry,
timeout=self.timeout,
metadata=self.metadata,
)
self.log.info("Endpoint was updated")
VertexAIEndpointLink.persist(context=context, task_instance=self, endpoint_id=self.endpoint_id)
return Endpoint.to_dict(result)