Method: tuningJobs.rebaseTunedModel

Full name: projects.locations.tuningJobs.rebaseTunedModel

Rebase a tuned model.

A rebase operation takes a model that was previously tuned on a base model version, and retunes it on a new base model version. The rebase operation creates a new tuning job and a new tuned model.

Endpoint

post https://aiplatform.googleapis.com/v1/{parent}/tuningJobs:rebaseTunedModel

Path parameters

parent string

Required. The resource name of the location in which to rebase the Model. Format: projects/{project}/locations/{location}

Request body

The request body contains data with the following structure:

Fields
tunedModelRef object (TunedModelRef)

Required. A reference to the tuned model to rebase.

tuningJob object (TuningJob)

Optional. The tuning job to be updated. Users can use this field to overwrite tuning configs.

artifactDestination object (GcsDestination)

Optional. The Google Cloud Storage location to write the artifacts to.

deployToSameEndpoint boolean

Optional. By default, rebasing a model creates a new endpoint for the new model. If this flag is set to true, the new model will be deployed to the same endpoint as the original model.

WARNING: If you deploy to the same endpoint, the original model will be un-deployed and replaced by the new model.

Response body

If successful, the response body contains an instance of Operation.

TunedModelRef

TunedModel Reference for legacy model migration.

Fields
tuned_model_ref Union type
The Tuned Model Reference for the model. tuned_model_ref can be only one of the following:
tunedModel string

Support migration from model registry.

tuningJob string

Support migration from tuning job list page, from gemini-1.0-pro-002 to 1.5 and above.

pipelineJob string

Support migration from tuning job list page, from bison model to gemini model.

JSON representation 
{

  // tuned_model_ref
  "tunedModel": string,
  "tuningJob": string,
  "pipelineJob": string
  // Union type
}