1   Definition

A Data Type that can be used to carry out Neural Network-based operations.

2   Functional Requirements

A NNModel is trained to enable the use of model to perform specific functions that might otherwise be performed by traditional methods.

3   Syntax


4   Semantics

Label Size Description
Header N1 Bytes Neural Network Model Header
– Standard-NeuralNetworkModel 9 Bytes The characters “AIF-NNM-V”
– Version N2 Bytes Major version – 1 or 2 characters
– Dot-separator 1 Byte The character “.”
– Subversion N3 Bytes Minor version – 1 or 2 characters
MInstanceID N4 Bytes Identifier of M-Instance.
NNModelID N5 Bytes Identifier of the Neural Network Model.
NNModelData N6 Bytes Standard set of Neural Network Model Data
– NNModelQualifier N7 Bytes Qualifier of Neural Network Model
– NNModelPayload N8 Bytes Set of Neural Network Model Data Length and URI
  – NNModelDataLength N9 Bytes Network Model Data Length in Bytes
  – NNModeDataURI N10 Bytes Network Model Data URI
DescrMetadata N10 Bytes Descriptive Metadata