Hi team,
I'm trying to understand how to use the offline_model attribute (specifically the artifact path) with this package for loading a pre-trained model locally rather than downloading it from a remote repository.
Could you please provide guidance or an example on how to properly configure and use this attribute?
I'm particularly interested in the following:
- The expected format of the artifact path
- How to integrate it with the existing API or model loading pipeline
- Any configuration or flags required to enable offline usage
Thanks in advance for your help!