gcloud ml-engine API

0
votes

Are gcloud ml-engine calls included in the google-cloud client library for python? I currently cannot find any documentation for it (though I see the natural language API). I am trying to replicate the following command in a jupyter notebook via the API and have not had any success:

gcloud ml-engine local predict --json-instances=XXX --model-dir=YYY

UPDATE w/ solution

with open('test.json') as data_file:    
    json_request = json.load(data_file)

response = predict_json(project = PROJECT_ID,
                        model= 'test_model',
                        instances = [json_request],
                        version = 'v1')
1
pythongoogle-cloud-platformgoogle-cloud-ml-engine

1 Answers

1
votes

I believe what you are looking for can be found in the official documentation under the section "Requesting predictions" (be sure to click on the Python tab).

For your convenience:

def predict_json(project, model, instances, version=None):
    """Send json data to a deployed model for prediction.

    Args:
        project (str): project where the Cloud ML Engine Model is deployed.
        model (str): model name.
        instances ([Mapping[str: Any]]): Keys should be the names of Tensors
            your deployed model expects as inputs. Values should be datatypes
            convertible to Tensors, or (potentially nested) lists of datatypes
            convertible to tensors.
        version: str, version of the model to target.
    Returns:
        Mapping[str: any]: dictionary of prediction results defined by the
            model.
    """
    # Create the ML Engine service object.
    # To authenticate set the environment variable
    # GOOGLE_APPLICATION_CREDENTIALS=<path_to_service_account_file>
    service = googleapiclient.discovery.build('ml', 'v1')
    name = 'projects/{}/models/{}'.format(project, model)

    if version is not None:
        name += '/versions/{}'.format(version)

    response = service.projects().predict(
        name=name,
        body={'instances': instances}
    ).execute()

    if 'error' in response:
        raise RuntimeError(response['error'])

    return response['predictions']