You are responsible for providing additional Python functions, API calls to reconstruct, deserialize your model into a form that you can work with.W&B suggests that you document information on how to load models into memory with model cards. For more information, see the Document machine learning models page.
<>
with your own:
latest
- Uselatest
alias to specify the model version that is most recently linked.v#
- Usev0
,v1
,v2
, and so on to fetch a specific version in the Registered Modelalias
- Specify the custom alias that you and your team assigned to your model version
use_model
in the API Reference guide for more information on possible parameters and return type.
Planned deprecation for W&B Model Registry in 2024The proceeding tabs demonstrate how to consume model artifacts using the soon to be deprecated Model Registry.Use the W&B Registry to track, organize and consume model artifacts. For more information see the Registry docs.
- CLI
- W&B App
Replace values within Reference a model version with one of following formats listed:
<>
with your own:latest
- Uselatest
alias to specify the model version that is most recently linked.v#
- Usev0
,v1
,v2
, and so on to fetch a specific version in the Registered Modelalias
- Specify the custom alias that you and your team assigned to your model version
use_model
in the API Reference guide for parameters and return type.