Package-level declarations
Types
The type of accelerator to use.
Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
Required. The type of goal to use for tuning. Available types are MAXIMIZE
and MINIMIZE
. Defaults to MAXIMIZE
.
metric name.
Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE
).
Required. The type of the parameter.
Required. The format of the input data files.
Optional. Format of the output data files, defaults to JSON.
The search algorithm specified for the study.
Required. Specifies the machine types, the number of replicas for workers and parameter servers.
Required. The optimization goal of the metric.
How the parameter should be scaled. Leave unset for categorical parameters.
Required. The type of the parameter.
The log type that this config enables.
The detailed state of a trial.
Optional. The machine learning framework AI Platform uses to train this version of the model. Valid values are TENSORFLOW
, SCIKIT_LEARN
, XGBOOST
. If you do not specify a framework, AI Platform will analyze files in the deployment_uri to determine a framework. If you choose SCIKIT_LEARN
or XGBOOST
, you must also set the runtime version of the model to 1.4 or greater. Do not specify a framework if you're deploying a /ai-platform/prediction/docs/custom-prediction-routines or if you're using a /ai-platform/prediction/docs/use-custom-container.