[
https://issues.apache.org/jira/browse/LUCENE-7179?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15226999#comment-15226999
]
Robert Muir commented on LUCENE-7179:
-------------------------------------
{quote}
But if we don't want to handle the dimensional reduction limitations we can
remove this approach altogether. Noting that we haven't even begun to tap into
its optimization potential.
{quote}
Right, I think we should try to improve it here to work the best it can in one
dimensional space. Working better on a z-order curve means doing work in
integer space! I have an example of what this looks like on LUCENE-7177. Look
in the code for all my comments about the encoding & rounding issues i worked
around.
If we can reason well about the encoding and quantization, we can explore
adapting that patch's 1D grid to work in the same z-order space, or other
options, or we can stick with using rectangles. Even if we stick with
rectangles, in my opinion integers and byte[] are easier for many operations,
so its worth the trouble.
> GeoPoint and LatLonPoint test data should quantize once
> -------------------------------------------------------
>
> Key: LUCENE-7179
> URL: https://issues.apache.org/jira/browse/LUCENE-7179
> Project: Lucene - Core
> Issue Type: Bug
> Reporter: Nicholas Knize
> Attachments: LUCENE-7179.patch
>
>
> {{LatLonPoint}} and {{GeoPointField}} tests pre quantizes test data to ensure
> consistency with indexed (encoded) data. The pre quantized data then becomes
> indexed, undergoing another quantization. To guarantee numerical stability
> this should be changed such that the test data is quantized after indexing.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
