Google
data class GoogleCloudMlV1__ParameterSpecArgs(val categoricalValues: Output<List<String>>? = null, val discreteValues: Output<List<Double>>? = null, val maxValue: Output<Double>? = null, val minValue: Output<Double>? = null, val parameterName: Output<String>, val scaleType: Output<GoogleCloudMlV1__ParameterSpecScaleType>? = null, val type: Output<GoogleCloudMlV1__ParameterSpecType>) : ConvertibleToJava<GoogleCloudMlV1__ParameterSpecArgs>
Represents a single hyperparameter to optimize.
Constructors
Link copied to clipboard
constructor(categoricalValues: Output<List<String>>? = null, discreteValues: Output<List<Double>>? = null, maxValue: Output<Double>? = null, minValue: Output<Double>? = null, parameterName: Output<String>, scaleType: Output<GoogleCloudMlV1__ParameterSpecScaleType>? = null, type: Output<GoogleCloudMlV1__ParameterSpecType>)
Properties
Link copied to clipboard
Required if type is CATEGORICAL. The list of possible categories.
Link copied to clipboard
Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
Link copied to clipboard
The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
Link copied to clipboard
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).
Link copied to clipboard
The type of the parameter.