topic Re: What is the best method for bringing an already trained model into MLflow? in Data Engineering

What is the best method for bringing an already trained model into MLflow?

User16826992666 — Fri, 25 Jun 2021 17:38:31 GMT

I already have a trained and saved model that was created outside of MLflow. What is the best way to handle it if I want this model to be added to an MLflow experiment?

Re: What is the best method for bringing an already trained model into MLflow?

Ryan_Chynoweth — Fri, 30 Jul 2021 16:09:51 GMT

If you can get the model into cloud storage and access the model via DBFS then you should be able to load the model as an object. Once it is loaded you can register it in MLflow or use it in an experiment run.

Re: What is the best method for bringing an already trained model into MLflow?

User16783853898 — Mon, 07 Mar 2022 16:35:14 GMT

https://databricks.atlassian.net/wiki/spaces/UN/pages/1280411724/MLflow+FAQ#How-do-I-create-an-MLflow-run-from-a-model-I-have-trained-elsewhere?

Re: What is the best method for bringing an already trained model into MLflow?

Anonymous — Fri, 22 Apr 2022 14:11:52 GMT

Hi @Trevor Bishop

Just wanted to check in if you were able to resolve your issue or do you need more help? We'd love to hear from you.

Thanks!