课程: MLOps with Databricks

免费学习该课程！

今天就开通帐号，24,600 门业界名师课程任您挑！

Set up, analyze, and process the inference table

Set up, analyze, and process the inference table - Databricks教程

课程: MLOps with Databricks

开始一个月试用 购买团队方案

Set up, analyze, and process the inference table

“

- [Instructor] Let's set up an inference table for the endpoint called hotel-cancellations-model-fe. I have already done that, but I will show you how you can do that as well. You can go to the edit button in the UI and scroll down. Go to the inference tables, click enable inference tables. Pass your catalog name, your schema name and the table name, and click on update. That can also be done programmatically, but I find that the easiest way. After the inference table is enabled, we can start sending requests to the end point. I already sent some requests before and you can see that most of the response times were under 50 milliseconds, but some took a bit longer. Let me show you how I send the request to the endpoint. We take the data from the test set and the data from the extra set that has different types of users. It's important not to forget to update the feature table and the online table with the data from the extra set. If you don't do that, our request to the endpoint will…

内容