Scheduling method and device of artificial intelligence model, equipment and storage medium
The invention provides an artificial intelligence model scheduling method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the steps of obtaining a calling request; the calling request comprises a to-be-completed target t...
Gespeichert in:
Hauptverfasser: | , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention provides an artificial intelligence model scheduling method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the steps of obtaining a calling request; the calling request comprises a to-be-completed target task; determining a target artificial intelligence model according to the calling request; the target artificial intelligence model is a model capable of completing the target task; the target artificial intelligence model is deployed in a preset container, the preset container is deployed in a graphics processor, and a plurality of preset containers can be deployed on the graphics processor; calling target monitoring information for the target artificial intelligence model, and determining a target model copy according to the target monitoring information; the target monitoring information comprises service state information of each model copy of the target artificial intelligence model and resource occupation info |
---|