Hi everyone, is there a best practice case can be shared to work with mlflow pyfunc in kedro?

Someone will reply to you shortly. In the meantime, this might help:

I'm not to familiar with pyfunc, @Dmitry Sorokin I know you did some work on mlflow do you any thoughts?

hi @Shu-Chun Wu , as I understood, you just need to specify

flavor: mlflow.pyfunc

in kedro catalog when describing your dataset, the rest should be according that manual in our docs: https://docs.kedro.org/en/stable/integrations/mlflow.html#artifact-tracking-in-mlflow-using-kedro-mlflow