AutoMLDeployModelOperator

Google

Deploys a model. If a model is already deployed, deploying it with the same parameters has no effect. Deploying with different parameters (as e.g. changing node_number) will reset the deployment state without pausing the model_id’s availability.

View on GitHub

Last Updated: Feb. 25, 2023

Access Instructions

Install the Google provider package into your Airflow environment.

Import the module into your DAG file and instantiate it with your desired params.

Parameters

Documentation

Connections

Parameters

model_idRequiredName of the model to be deployed.

image_detection_metadataModel deployment metadata specific to Image Object Detection. If a dict is provided, it must be of the same form as the protobuf message ImageObjectDetectionModelDeploymentMetadata

project_idID of the Google Cloud project where model is located if None then default project_id is used.

locationRequiredThe location of the project.

paramsAdditional domain-specific parameters for the predictions.

retryA retry object used to retry requests. If None is specified, requests will not be retried.

timeoutThe amount of time, in seconds, to wait for the request to complete. Note that if retry is specified, the timeout applies to each individual attempt.

metadataAdditional metadata that is provided to the method.

gcp_conn_idThe connection ID to use to connect to Google Cloud.

impersonation_chainOptional 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).

Documentation

Only applicable for Text Classification, Image Object Detection and Tables; all other domains manage deployment automatically.