| 1 | # Generated by default/object.tt | ||
| 2 | package Paws::Forecast::FeaturizationConfig; | ||
| 3 | use Moose; | ||
| 4 | has Featurizations => (is => 'ro', isa => 'ArrayRef[Paws::Forecast::Featurization]'); | ||
| 5 | has ForecastDimensions => (is => 'ro', isa => 'ArrayRef[Str|Undef]'); | ||
| 6 | has ForecastFrequency => (is => 'ro', isa => 'Str', required => 1); | ||
| 7 | |||
| 8 | 1; | ||
| 9 | |||
| 10 | ### main pod documentation begin ### | ||
| 11 | |||
| 12 | =head1 NAME | ||
| 13 | |||
| 14 | Paws::Forecast::FeaturizationConfig | ||
| 15 | |||
| 16 | =head1 USAGE | ||
| 17 | |||
| 18 | This class represents one of two things: | ||
| 19 | |||
| 20 | =head3 Arguments in a call to a service | ||
| 21 | |||
| 22 | Use the attributes of this class as arguments to methods. You shouldn't make instances of this class. | ||
| 23 | Each attribute should be used as a named argument in the calls that expect this type of object. | ||
| 24 | |||
| 25 | As an example, if Att1 is expected to be a Paws::Forecast::FeaturizationConfig object: | ||
| 26 | |||
| 27 | $service_obj->Method(Att1 => { Featurizations => $value, ..., ForecastFrequency => $value }); | ||
| 28 | |||
| 29 | =head3 Results returned from an API call | ||
| 30 | |||
| 31 | Use accessors for each attribute. If Att1 is expected to be an Paws::Forecast::FeaturizationConfig object: | ||
| 32 | |||
| 33 | $result = $service_obj->Method(...); | ||
| 34 | $result->Att1->Featurizations | ||
| 35 | |||
| 36 | =head1 DESCRIPTION | ||
| 37 | |||
| 38 | In a CreatePredictor operation, the specified algorithm trains a model | ||
| 39 | using the specified dataset group. You can optionally tell the | ||
| 40 | operation to modify data fields prior to training a model. These | ||
| 41 | modifications are referred to as I<featurization>. | ||
| 42 | |||
| 43 | You define featurization using the C<FeaturizationConfig> object. You | ||
| 44 | specify an array of transformations, one for each field that you want | ||
| 45 | to featurize. You then include the C<FeaturizationConfig> object in | ||
| 46 | your C<CreatePredictor> request. Amazon Forecast applies the | ||
| 47 | featurization to the C<TARGET_TIME_SERIES> and C<RELATED_TIME_SERIES> | ||
| 48 | datasets before model training. | ||
| 49 | |||
| 50 | You can create multiple featurization configurations. For example, you | ||
| 51 | might call the C<CreatePredictor> operation twice by specifying | ||
| 52 | different featurization configurations. | ||
| 53 | |||
| 54 | =head1 ATTRIBUTES | ||
| 55 | |||
| 56 | |||
| 57 | =head2 Featurizations => ArrayRef[L<Paws::Forecast::Featurization>] | ||
| 58 | |||
| 59 | An array of featurization (transformation) information for the fields | ||
| 60 | of a dataset. | ||
| 61 | |||
| 62 | |||
| 63 | =head2 ForecastDimensions => ArrayRef[Str|Undef] | ||
| 64 | |||
| 65 | An array of dimension (field) names that specify how to group the | ||
| 66 | generated forecast. | ||
| 67 | |||
| 68 | For example, suppose that you are generating a forecast for item sales | ||
| 69 | across all of your stores, and your dataset contains a C<store_id> | ||
| 70 | field. If you want the sales forecast for each item by store, you would | ||
| 71 | specify C<store_id> as the dimension. | ||
| 72 | |||
| 73 | All forecast dimensions specified in the C<TARGET_TIME_SERIES> dataset | ||
| 74 | don't need to be specified in the C<CreatePredictor> request. All | ||
| 75 | forecast dimensions specified in the C<RELATED_TIME_SERIES> dataset | ||
| 76 | must be specified in the C<CreatePredictor> request. | ||
| 77 | |||
| 78 | |||
| 79 | =head2 B<REQUIRED> ForecastFrequency => Str | ||
| 80 | |||
| 81 | The frequency of predictions in a forecast. | ||
| 82 | |||
| 83 | Valid intervals are Y (Year), M (Month), W (Week), D (Day), H (Hour), | ||
| 84 | 30min (30 minutes), 15min (15 minutes), 10min (10 minutes), 5min (5 | ||
| 85 | minutes), and 1min (1 minute). For example, "Y" indicates every year | ||
| 86 | and "5min" indicates every five minutes. | ||
| 87 | |||
| 88 | The frequency must be greater than or equal to the TARGET_TIME_SERIES | ||
| 89 | dataset frequency. | ||
| 90 | |||
| 91 | When a RELATED_TIME_SERIES dataset is provided, the frequency must be | ||
| 92 | equal to the RELATED_TIME_SERIES dataset frequency. | ||
| 93 | |||
| 94 | |||
| 95 | |||
| 96 | =head1 SEE ALSO | ||
| 97 | |||
| 98 | This class forms part of L<Paws>, describing an object used in L<Paws::Forecast> | ||
| 99 | |||
| 100 | =head1 BUGS and CONTRIBUTIONS | ||
| 101 | |||
| 102 | The source code is located here: L<https://github.com/pplu/aws-sdk-perl> | ||
| 103 | |||
| 104 | Please report bugs to: L<https://github.com/pplu/aws-sdk-perl/issues> | ||
| 105 | |||
| 106 | =cut | ||
| 107 |