Amazonka.SageMaker.Types.Channel

Documentation

data Channel

A channel is a named input source that training algorithms can consume.

See: newChannel smart constructor.

Constructors

Channel'

Fields

• compressionType :: Maybe CompressionType

  If training data is compressed, the compression type. The default value is None. CompressionType is used only in Pipe input mode. In File mode, leave this field unset or set it to None.

• contentType :: Maybe Text

  The MIME type of the data.

• inputMode :: Maybe TrainingInputMode

  (Optional) The input mode to use for the data channel in a training job. If you don't set a value for InputMode, SageMaker uses the value set for TrainingInputMode. Use this parameter to override the TrainingInputMode setting in a AlgorithmSpecification request when you have a channel that needs a different input mode from the training job's general setting. To download the data from Amazon Simple Storage Service (Amazon S3) to the provisioned ML storage volume, and mount the directory to a Docker volume, use File input mode. To stream data directly from Amazon S3 to the container, choose Pipe input mode.

  To use a model for incremental training, choose File input model.

• recordWrapperType :: Maybe RecordWrapper

  Specify RecordIO as the value when input data is in raw format but the training algorithm requires the RecordIO format. In this case, SageMaker wraps each individual S3 object in a RecordIO record. If the input data is already in RecordIO format, you don't need to set this attribute. For more information, see Create a Dataset Using RecordIO.

  In File mode, leave this field unset or set it to None.

• shuffleConfig :: Maybe ShuffleConfig

  A configuration for a shuffle option for input data in a channel. If you use S3Prefix for S3DataType, this shuffles the results of the S3 key prefix matches. If you use ManifestFile, the order of the S3 object references in the ManifestFile is shuffled. If you use AugmentedManifestFile, the order of the JSON lines in the AugmentedManifestFile is shuffled. The shuffling order is determined using the Seed value.

  For Pipe input mode, shuffling is done at the start of every epoch. With large datasets this ensures that the order of the training data is different for each epoch, it helps reduce bias and possible overfitting. In a multi-node training job when ShuffleConfig is combined with S3DataDistributionType of ShardedByS3Key, the data is shuffled across nodes so that the content sent to a particular node on the first epoch might be sent to a different node on the second epoch.

• channelName :: Text

  The name of the channel.

• dataSource :: DataSource

  The location of the channel data.

Instances

FromJSON Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

parseJSON :: Value -> Parser Channel

parseJSONList :: Value -> Parser [Channel]

ToJSON Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

toJSON :: Channel -> Value

toEncoding :: Channel -> Encoding

toJSONList :: [Channel] -> Value

toEncodingList :: [Channel] -> Encoding

Generic Channel

Defined in Amazonka.SageMaker.Types.Channel

Associated Types

type Rep Channel :: Type -> Type

Methods

from :: Channel -> Rep Channel x

to :: Rep Channel x -> Channel

Read Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

readsPrec :: Int -> ReadS Channel

readList :: ReadS [Channel]

readPrec :: ReadPrec Channel

readListPrec :: ReadPrec [Channel]

Show Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

showsPrec :: Int -> Channel -> ShowS

show :: Channel -> String

showList :: [Channel] -> ShowS

NFData Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

rnf :: Channel -> ()

Eq Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

(==) :: Channel -> Channel -> Bool

(/=) :: Channel -> Channel -> Bool

Hashable Channel

Defined in Amazonka.SageMaker.Types.Channel

Methods

hashWithSalt :: Int -> Channel -> Int

hash :: Channel -> Int

type Rep Channel

Defined in Amazonka.SageMaker.Types.Channel

type Rep Channel = D1 ('MetaData "Channel" "Amazonka.SageMaker.Types.Channel" "amazonka-sagemaker-2.0-9SyrKZ4KqhsL1qX9u3ILA3" 'False) (C1 ('MetaCons "Channel'" 'PrefixI 'True) ((S1 ('MetaSel ('Just "compressionType") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe CompressionType)) :*: (S1 ('MetaSel ('Just "contentType") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe Text)) :*: S1 ('MetaSel ('Just "inputMode") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe TrainingInputMode)))) :*: ((S1 ('MetaSel ('Just "recordWrapperType") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe RecordWrapper)) :*: S1 ('MetaSel ('Just "shuffleConfig") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe ShuffleConfig))) :*: (S1 ('MetaSel ('Just "channelName") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 Text) :*: S1 ('MetaSel ('Just "dataSource") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 DataSource)))))

newChannel

Arguments

:: Text

$sel:channelName:Channel'

-> DataSource

$sel:dataSource:Channel'

-> Channel

Create a value of Channel with all optional fields omitted.

Use generic-lens or optics to modify other optional fields.

The following record fields are available, with the corresponding lenses provided for backwards compatibility:

$sel:compressionType:Channel', channel_compressionType - If training data is compressed, the compression type. The default value is None. CompressionType is used only in Pipe input mode. In File mode, leave this field unset or set it to None.

$sel:contentType:Channel', channel_contentType - The MIME type of the data.

$sel:inputMode:Channel', channel_inputMode - (Optional) The input mode to use for the data channel in a training job. If you don't set a value for InputMode, SageMaker uses the value set for TrainingInputMode. Use this parameter to override the TrainingInputMode setting in a AlgorithmSpecification request when you have a channel that needs a different input mode from the training job's general setting. To download the data from Amazon Simple Storage Service (Amazon S3) to the provisioned ML storage volume, and mount the directory to a Docker volume, use File input mode. To stream data directly from Amazon S3 to the container, choose Pipe input mode.

To use a model for incremental training, choose File input model.

$sel:recordWrapperType:Channel', channel_recordWrapperType - Specify RecordIO as the value when input data is in raw format but the training algorithm requires the RecordIO format. In this case, SageMaker wraps each individual S3 object in a RecordIO record. If the input data is already in RecordIO format, you don't need to set this attribute. For more information, see Create a Dataset Using RecordIO.

In File mode, leave this field unset or set it to None.

$sel:shuffleConfig:Channel', channel_shuffleConfig - A configuration for a shuffle option for input data in a channel. If you use S3Prefix for S3DataType, this shuffles the results of the S3 key prefix matches. If you use ManifestFile, the order of the S3 object references in the ManifestFile is shuffled. If you use AugmentedManifestFile, the order of the JSON lines in the AugmentedManifestFile is shuffled. The shuffling order is determined using the Seed value.

For Pipe input mode, shuffling is done at the start of every epoch. With large datasets this ensures that the order of the training data is different for each epoch, it helps reduce bias and possible overfitting. In a multi-node training job when ShuffleConfig is combined with S3DataDistributionType of ShardedByS3Key, the data is shuffled across nodes so that the content sent to a particular node on the first epoch might be sent to a different node on the second epoch.

$sel:channelName:Channel', channel_channelName - The name of the channel.

$sel:dataSource:Channel', channel_dataSource - The location of the channel data.

channel_compressionType :: Lens' Channel (Maybe CompressionType)

If training data is compressed, the compression type. The default value is None. CompressionType is used only in Pipe input mode. In File mode, leave this field unset or set it to None.

channel_contentType :: Lens' Channel (Maybe Text)

The MIME type of the data.

channel_inputMode :: Lens' Channel (Maybe TrainingInputMode)

(Optional) The input mode to use for the data channel in a training job. If you don't set a value for InputMode, SageMaker uses the value set for TrainingInputMode. Use this parameter to override the TrainingInputMode setting in a AlgorithmSpecification request when you have a channel that needs a different input mode from the training job's general setting. To download the data from Amazon Simple Storage Service (Amazon S3) to the provisioned ML storage volume, and mount the directory to a Docker volume, use File input mode. To stream data directly from Amazon S3 to the container, choose Pipe input mode.

To use a model for incremental training, choose File input model.

channel_recordWrapperType :: Lens' Channel (Maybe RecordWrapper)

Specify RecordIO as the value when input data is in raw format but the training algorithm requires the RecordIO format. In this case, SageMaker wraps each individual S3 object in a RecordIO record. If the input data is already in RecordIO format, you don't need to set this attribute. For more information, see Create a Dataset Using RecordIO.

In File mode, leave this field unset or set it to None.

channel_shuffleConfig :: Lens' Channel (Maybe ShuffleConfig)

A configuration for a shuffle option for input data in a channel. If you use S3Prefix for S3DataType, this shuffles the results of the S3 key prefix matches. If you use ManifestFile, the order of the S3 object references in the ManifestFile is shuffled. If you use AugmentedManifestFile, the order of the JSON lines in the AugmentedManifestFile is shuffled. The shuffling order is determined using the Seed value.

For Pipe input mode, shuffling is done at the start of every epoch. With large datasets this ensures that the order of the training data is different for each epoch, it helps reduce bias and possible overfitting. In a multi-node training job when ShuffleConfig is combined with S3DataDistributionType of ShardedByS3Key, the data is shuffled across nodes so that the content sent to a particular node on the first epoch might be sent to a different node on the second epoch.

channel_channelName :: Lens' Channel Text

The name of the channel.

channel_dataSource :: Lens' Channel DataSource

The location of the channel data.