« blib/lib/Paws/SageMaker/HyperParameterTuningJobSummary.pm
↑ Summary
blib/lib/Paws/SageMaker/HyperParameters.pm »
| 1 | # Generated by default/object.tt | ||
| 2 | package Paws::SageMaker::HyperParameterTuningJobWarmStartConfig; | ||
| 3 | use Moose; | ||
| 4 | has ParentHyperParameterTuningJobs => (is => 'ro', isa => 'ArrayRef[Paws::SageMaker::ParentHyperParameterTuningJob]', required => 1); | ||
| 5 | has WarmStartType => (is => 'ro', isa => 'Str', required => 1); | ||
| 6 | |||
| 7 | 1; | ||
| 8 | |||
| 9 | ### main pod documentation begin ### | ||
| 10 | |||
| 11 | =head1 NAME | ||
| 12 | |||
| 13 | Paws::SageMaker::HyperParameterTuningJobWarmStartConfig | ||
| 14 | |||
| 15 | =head1 USAGE | ||
| 16 | |||
| 17 | This class represents one of two things: | ||
| 18 | |||
| 19 | =head3 Arguments in a call to a service | ||
| 20 | |||
| 21 | Use the attributes of this class as arguments to methods. You shouldn't make instances of this class. | ||
| 22 | Each attribute should be used as a named argument in the calls that expect this type of object. | ||
| 23 | |||
| 24 | As an example, if Att1 is expected to be a Paws::SageMaker::HyperParameterTuningJobWarmStartConfig object: | ||
| 25 | |||
| 26 | $service_obj->Method(Att1 => { ParentHyperParameterTuningJobs => $value, ..., WarmStartType => $value }); | ||
| 27 | |||
| 28 | =head3 Results returned from an API call | ||
| 29 | |||
| 30 | Use accessors for each attribute. If Att1 is expected to be an Paws::SageMaker::HyperParameterTuningJobWarmStartConfig object: | ||
| 31 | |||
| 32 | $result = $service_obj->Method(...); | ||
| 33 | $result->Att1->ParentHyperParameterTuningJobs | ||
| 34 | |||
| 35 | =head1 DESCRIPTION | ||
| 36 | |||
| 37 | Specifies the configuration for a hyperparameter tuning job that uses | ||
| 38 | one or more previous hyperparameter tuning jobs as a starting point. | ||
| 39 | The results of previous tuning jobs are used to inform which | ||
| 40 | combinations of hyperparameters to search over in the new tuning job. | ||
| 41 | |||
| 42 | All training jobs launched by the new hyperparameter tuning job are | ||
| 43 | evaluated by using the objective metric, and the training job that | ||
| 44 | performs the best is compared to the best training jobs from the parent | ||
| 45 | tuning jobs. From these, the training job that performs the best as | ||
| 46 | measured by the objective metric is returned as the overall best | ||
| 47 | training job. | ||
| 48 | |||
| 49 | All training jobs launched by parent hyperparameter tuning jobs and the | ||
| 50 | new hyperparameter tuning jobs count against the limit of training jobs | ||
| 51 | for the tuning job. | ||
| 52 | |||
| 53 | =head1 ATTRIBUTES | ||
| 54 | |||
| 55 | |||
| 56 | =head2 B<REQUIRED> ParentHyperParameterTuningJobs => ArrayRef[L<Paws::SageMaker::ParentHyperParameterTuningJob>] | ||
| 57 | |||
| 58 | An array of hyperparameter tuning jobs that are used as the starting | ||
| 59 | point for the new hyperparameter tuning job. For more information about | ||
| 60 | warm starting a hyperparameter tuning job, see Using a Previous | ||
| 61 | Hyperparameter Tuning Job as a Starting Point | ||
| 62 | (https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-warm-start.html). | ||
| 63 | |||
| 64 | Hyperparameter tuning jobs created before October 1, 2018 cannot be | ||
| 65 | used as parent jobs for warm start tuning jobs. | ||
| 66 | |||
| 67 | |||
| 68 | =head2 B<REQUIRED> WarmStartType => Str | ||
| 69 | |||
| 70 | Specifies one of the following: | ||
| 71 | |||
| 72 | =over | ||
| 73 | |||
| 74 | =item IDENTICAL_DATA_AND_ALGORITHM | ||
| 75 | |||
| 76 | The new hyperparameter tuning job uses the same input data and training | ||
| 77 | image as the parent tuning jobs. You can change the hyperparameter | ||
| 78 | ranges to search and the maximum number of training jobs that the | ||
| 79 | hyperparameter tuning job launches. You cannot use a new version of the | ||
| 80 | training algorithm, unless the changes in the new version do not affect | ||
| 81 | the algorithm itself. For example, changes that improve logging or | ||
| 82 | adding support for a different data format are allowed. You can also | ||
| 83 | change hyperparameters from tunable to static, and from static to | ||
| 84 | tunable, but the total number of static plus tunable hyperparameters | ||
| 85 | must remain the same as it is in all parent jobs. The objective metric | ||
| 86 | for the new tuning job must be the same as for all parent jobs. | ||
| 87 | |||
| 88 | =item TRANSFER_LEARNING | ||
| 89 | |||
| 90 | The new hyperparameter tuning job can include input data, | ||
| 91 | hyperparameter ranges, maximum number of concurrent training jobs, and | ||
| 92 | maximum number of training jobs that are different than those of its | ||
| 93 | parent hyperparameter tuning jobs. The training image can also be a | ||
| 94 | different version from the version used in the parent hyperparameter | ||
| 95 | tuning job. You can also change hyperparameters from tunable to static, | ||
| 96 | and from static to tunable, but the total number of static plus tunable | ||
| 97 | hyperparameters must remain the same as it is in all parent jobs. The | ||
| 98 | objective metric for the new tuning job must be the same as for all | ||
| 99 | parent jobs. | ||
| 100 | |||
| 101 | =back | ||
| 102 | |||
| 103 | |||
| 104 | |||
| 105 | |||
| 106 | =head1 SEE ALSO | ||
| 107 | |||
| 108 | This class forms part of L<Paws>, describing an object used in L<Paws::SageMaker> | ||
| 109 | |||
| 110 | =head1 BUGS and CONTRIBUTIONS | ||
| 111 | |||
| 112 | The source code is located here: L<https://github.com/pplu/aws-sdk-perl> | ||
| 113 | |||
| 114 | Please report bugs to: L<https://github.com/pplu/aws-sdk-perl/issues> | ||
| 115 | |||
| 116 | =cut | ||
| 117 |