Glossary term
Glossary term
Infrastructure and Serving
Describes the information required to extract features data from the tf.Example protocol buffer. Because the tf.Example protocol buffer is just a container for data, you must specify the following:
The data to extract (that is, the keys for the features)
The data type (for example, float or int)
The length (fixed or variable)
Created for this library
A data engineering team defines a feature spec per feature, including type and acceptable range, so production data quality is enforced at ingestion.
A risk modeling team's feature spec includes a description of business meaning so model auditors can interpret each input without reading code.
A churn team uses feature spec metadata to drive its training and serving pipelines from a single source of truth.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License