xudong963 commented on PR #15296: URL: https://github.com/apache/datafusion/pull/15296#issuecomment-2743593315
> Do you know any use cases where this method would be especially useful? If so, maybe we can study one of those cases in more detail. That could help us understand the real need and guide us toward a more solid algorithm. Yes, we're considering restarting the [work](https://github.com/apache/datafusion/pull/13296/files#diff-8d786f45bc2d5bf629754a119ed6fa7998dcff7faacd954c45945b7047b87fa1R498), and given that `Precision` will be replaced with `Distribution`, so I opened the proposal to discuss how to do **merge** for `Distribution`. > a) Mixture Model (Weighted Average of PDFs) > This is a method for combining different probability distributions. > p1(x) and p2(x) is some PDF's, and we give them equal weight (0.5). The combined PDF would be: > _pmix(x) = 0.5 * p1(x) + 0.5 * p2(x)_ > This creates a probabilistic blend of the two. The result is still a valid PDF (non-negative and integrates to 1). It seems that my current way is close to the mixture model. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
