Certain first-party algorithms (like K-Means, PCA, and LinearLearner) on Amazon SageMaker expect each Amazon Record protobuf message to contain a "features" map with key "values" that encodes the example's features vector.
Converts a byte array to a Protobuf Record.
Converts a byte array to a Protobuf Record.
byte array to convert to Protobuf Record
Protobuf Record from byte array
Encodes a byte array in recordIO.
Encodes a byte array in recordIO.
Byte array to encode in recordIO
RecordIO encoded byte array
https://mxnet.incubator.apache.org/architecture/note_data_loading.html for more information on recordIO
Converts a recordIO encoded byte array of Protobuf records to an iterator of Protobuf Records
Converts a recordIO encoded byte array of Protobuf records to an iterator of Protobuf Records
recordIO encoded byte array to convert to Protobuf Record
An Iterator over Protobuf Records
https://mxnet.incubator.apache.org/architecture/note_data_loading.html for more information on recordIO
Converts a Row to a SageMaker Protobuf record for training or inference.
Converts a Row to a SageMaker Protobuf record for training or inference.
a Row with a column of Doubles for labels and a Vector for features
index in the row corresponding to the features column
index in the row corresponding to the label column
a SageMaker Protobuf record representing the Row
Utility functions that convert to and from the Amazon Record protobuf data format and encode Records in recordIO.
https://mxnet.incubator.apache.org/architecture/note_data_loading.html for more information on recordIO
https://aws.amazon.com/sagemaker/latest/dg/cdf-training.html/ for more information on the Amazon Record data format