NAME

gcloud alpha ml-engine local predict - run prediction locally

SYNOPSIS

gcloud alpha ml-engine local predict --model-dir=MODEL_DIR (--json-instances=JSON_INSTANCES | --json-request=JSON_REQUEST | --text-instances=TEXT_INSTANCES) [--framework=FRAMEWORK] [--signature-name=SIGNATURE_NAME] [GCLOUD_WIDE_FLAG ...]

DESCRIPTION

(ALPHA) gcloud alpha ml-engine local predict performs prediction locally with the given instances. It requires the TensorFlow SDK https://www.tensorflow.org/install be installed locally. The output format mirrors gcloud ai-platform predict (online prediction).

You cannot use this command with custom prediction routines.

REQUIRED FLAGS

--model-dir=MODEL_DIR

Path to the model.

Exactly one of these must be specified:
--json-instances=JSON_INSTANCES

Path to a local file from which instances are read. Instances are in JSON format; newline delimited.

An example of the JSON instances file:

{"images": [0.0, ..., 0.1], "key": 3} {"images": [0.0, ..., 0.1], "key": 2}

This flag accepts "-" for stdin.

--json-request=JSON_REQUEST

Path to a local file containing the body of JSON request.

An example of a JSON request:

{ "instances": [ {"x": [1, 2], "y": [3, 4]}, {"x": [-1, -2], "y": [-3, -4]} ] }

This flag accepts "-" for stdin.

--text-instances=TEXT_INSTANCES

Path to a local file from which instances are read. Instances are in UTF-8 encoded text format; newline delimited.

An example of the text instances file:

107,4.9,2.5,4.5,1.7 100,5.7,2.8,4.1,1.3

This flag accepts "-" for stdin.

OPTIONAL FLAGS

--framework=FRAMEWORK

ML framework used to train this version of the model. If not specified, defaults to 'tensorflow'. FRAMEWORK must be one of: scikit-learn, tensorflow, xgboost.

--signature-name=SIGNATURE_NAME

Name of the signature defined in the SavedModel to use for this job. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY in https://www.tensorflow.org/api_docs/python/tf/compat/v1/saved_model/signature_constants, which is "serving_default". Only applies to TensorFlow models.

GCLOUD WIDE FLAGS

These flags are available to all commands: --access-token-file, --account, --billing-project, --configuration, --flags-file, --flatten, --format, --help, --impersonate-service-account, --log-http, --project, --quiet, --trace-token, --user-output-enabled, --verbosity.

Run $ gcloud help for details.

NOTES

This command is currently in alpha and might change without notice. If this command fails with API permission errors despite specifying the correct project, you might be trying to access an API with an invitation-only early access allowlist. These variants are also available:

$ gcloud ml-engine local predict

$ gcloud beta ml-engine local predict