by
admo
• New Contributor III
- 9555 Views
- 4 replies
- 7 kudos
Hello,I'm writing this because I have tried a lot of different directions to get a simple model inference working with no success.Here is the outline of the job# 1 - Load the base data (~1 billion lines of ~6 columns)
interaction = build_initial_df()...
- 9555 Views
- 4 replies
- 7 kudos
Latest Reply
It is hard to analyze without Spark UI and more detailed information, but anyway few tips:look for data skews some partitions can be very big some small because of incorrect partitioning. You can use Spark UI to do that but also debug your code a bit...
3 More Replies
- 1770 Views
- 1 replies
- 0 kudos
What kind of latency should I expect when using the built in model serving capability in MLflow. Evaluating whether it would be a good fit for our use case
- 1770 Views
- 1 replies
- 0 kudos
Latest Reply
What are your throughput requirements in addition to latency. Currently this is in private preview and databricks recommends this only for low throughput and non-critical applications. However, as it move towards GA, this would change. Please get in...
- 1508 Views
- 0 replies
- 0 kudos
hp.quniform (“quantized uniform”) or hp.qloguniform to generate integers. hp.choice is the right choice when, for example, choosing among categorical choices (which might in some situations even be integers, but not usually).https://databricks.com/b...
- 1508 Views
- 0 replies
- 0 kudos
- 4199 Views
- 3 replies
- 1 kudos
They both seem to package it. When should one use one over the other?
- 4199 Views
- 3 replies
- 1 kudos
Latest Reply
One thing I think it's useful to point out for Databricks users is that you would typically not use MLflow Projects to describe execution of a modeling run. You would just use MLflow directly in Databricks and use Databricks notebooks to manage code ...
2 More Replies
