You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I need to save the model checkpoint trained using TabularModel. I wish to use this model checkpoint later into an external library called "alibi-detect". It is useful for detecting data drift statistically.
You can save the model as a pure PyTorch model. But you will have to handle the data transformations like normalisation etc on your own by saving the datamodule separately.
#268 has details on how to save models in different formats
I need to save the model checkpoint trained using TabularModel. I wish to use this model checkpoint later into an external library called "alibi-detect". It is useful for detecting data drift statistically.
Alibi-detect documentation states it supports loading ML models saved as per the link : https://docs.seldon.io/projects/alibi-detect/en/stable/overview/saving.html#supported-ml-models
Is there a way to load a TabularModel into Alibi-Detect detector?
The text was updated successfully, but these errors were encountered: