RecordSet: RecordSet Class

In DyfanJones/sagemaker-r-mlcore: sagemaker machine learning core classes and methods

RecordSet Class

Description

RecordSet Class

RecordSet Class

Methods

Public methods

• RecordSet$new()

• RecordSet$data_channel()

• RecordSet$records_s3_input()

• RecordSet$print()

• RecordSet$clone()

---

Method new()

A collection of Amazon :class:~'Record' objects serialized and stored in S3.

Usage

RecordSet$new(
  s3_data,
  num_records,
  feature_dim,
  s3_data_type = "ManifestFile",
  channel = "train"
)

Arguments

s3_data

(str): The S3 location of the training data

num_records

(int): The number of records in the set.

feature_dim

(int): The dimensionality of "values" arrays in the Record features, and label (if each Record is labeled).

s3_data_type

(str): Valid values: 'S3Prefix', 'ManifestFile'. If 'S3Prefix', “s3_data“ defines a prefix of s3 objects to train on. All objects with s3 keys beginning with “s3_data“ will be used to train. If 'ManifestFile', then “s3_data“ defines a single s3 manifest file, listing each s3 object to train on.

channel

(str): The SageMaker Training Job channel this RecordSet should be bound to

---

Method data_channel()

Return a dictionary to represent the training data in a channel for use with “fit()“

Usage

RecordSet$data_channel()

---

Method records_s3_input()

Return a TrainingInput to represent the training data

Usage

RecordSet$records_s3_input()

---

Method print()

Return an unambiguous representation of this RecordSet

Usage

RecordSet$print()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

RecordSet$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

DyfanJones/sagemaker-r-mlcore documentation built on May 3, 2022, 10:08 a.m.