<- Go to Data Types Go to ToC

1. Definition 2. Functional Requirements 3. Syntax 4. Semantics

1 Definition

ML Model Object refers to

  1. ML Model Data an instance of which results from the application of training data to a process.
  2. ML Model Qualifier specified by MPAI-TFA providing information on Sub-Types, Formats and Attributes.

2 Functional Requirements

A Machine Learning Model enables the performance of specific functions most of which are target of MPAI Technical Specifications, e.g., classification.

3 Syntax

https://schemas.mpai.community/AIF/V2.2/data/MLModelObject.json

4 Semantics

Label Description
Header ML Model Header
– Standard-MLModelObject The characters “AIF-MLO-V”
– Version Major version – 1 or 2 characters
– Dot-separator The character “.”
– Subversion Minor version – 1 or 2 characters
MInstanceID Identifier of M-Instance.
MLModelObjectID Identifier of Machine Learning Model.
MLModelData The set of data constituting the ML Model.
– DataLength and DataURI and/or The number of bytes of the MLModelDataand the URI and/or
– MLModelData and/or The MLModelData and/or
– MLModelDataID MLModelDataID.
MLModelQualifier Qualifier of ML Model.
DescrMetadata Descriptive Metadata.

<- Go to Data Types Go to ToC