Class: Aws::SageMaker::Types::ShuffleConfig

Inherits:

Struct

• Object
• Struct
• Aws::SageMaker::Types::ShuffleConfig

Defined in:

(unknown)

Overview

Note:

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

{
  seed: 1, # required
}

A configuration for a shuffle option for input data in a channel. If you use S3Prefix for S3DataType, the results of the S3 key prefix matches are shuffled. 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, when ShuffleConfig is specified 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, and 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.

Returned by:

• Channel#shuffle_config

See Also:

• AWS API Documentation

Instance Attribute Summary

• #seed ⇒ Integer

  Determines the shuffling order in ShuffleConfig value.

Instance Attribute Details

#seed ⇒ Integer

Determines the shuffling order in ShuffleConfig value.

Returns:

• (Integer) —

  Determines the shuffling order in ShuffleConfig value.