You are viewing documentation for version 2 of the AWS SDK for Ruby. Version 3 documentation can be found here.

Class: Aws::ForecastService::Types::HyperParameterTuningJobConfig

Inherits:
Struct
  • Object
show all
Defined in:
(unknown)

Overview

Note:

When passing HyperParameterTuningJobConfig as input to an Aws::Client method, you can use a vanilla Hash:

{
  parameter_ranges: {
    categorical_parameter_ranges: [
      {
        name: "Name", # required
        values: ["Value"], # required
      },
    ],
    continuous_parameter_ranges: [
      {
        name: "Name", # required
        max_value: 1.0, # required
        min_value: 1.0, # required
        scaling_type: "Auto", # accepts Auto, Linear, Logarithmic, ReverseLogarithmic
      },
    ],
    integer_parameter_ranges: [
      {
        name: "Name", # required
        max_value: 1, # required
        min_value: 1, # required
        scaling_type: "Auto", # accepts Auto, Linear, Logarithmic, ReverseLogarithmic
      },
    ],
  },
}

Configuration information for a hyperparameter tuning job. This object is specified in the CreatePredictor request.

A hyperparameter is a parameter that governs the model training process and is set before training starts. This is as opposed to a model parameter that is determined during training. The values of the hyperparameters have an effect on the chosen model parameters.

A hyperparameter tuning job is the process of choosing the optimum set of hyperparameter values that optimize a specified metric. This is accomplished by running many training jobs over a range of hyperparameter values. The optimum set of values is dependent on the algorithm, the training data, and the given metric objective.

Returned by:

Instance Attribute Summary collapse

Instance Attribute Details

#parameter_rangesTypes::ParameterRanges

Specifies the ranges of valid values for the hyperparameters.

Returns: