News for hackers. Stuff that matters.

Ask HN: How do you deploy your ML models in production?

I use Spark MLLib to train models.

In most of the cases, the predictions are meant to be delivered via a REST API. One of the approaches is to pre-compute predictions -> store in DB -> serve via API.

Do you have a better pipeline for your production models?

( no comments | 2 points | on HN )

There are no comments yet.