You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@solr.apache.org by "andywebb1975 (via GitHub)" <gi...@apache.org> on 2023/03/04 11:25:40 UTC
[GitHub] [solr] andywebb1975 opened a new pull request, #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
andywebb1975 opened a new pull request, #1431:
URL: https://github.com/apache/solr/pull/1431
https://issues.apache.org/jira/browse/SOLR-16643
<!--
_(If you are a project committer then you may remove some/all of the following template.)_
Before creating a pull request, please file an issue in the ASF Jira system for Solr:
* https://issues.apache.org/jira/projects/SOLR
For something minor (i.e. that wouldn't be worth putting in release notes), you can skip JIRA.
To create a Jira issue, you will need to create an account there first.
The title of the PR should reference the Jira issue number in the form:
* SOLR-####: <short description of problem or changes>
SOLR must be fully capitalized. A short description helps people scanning pull requests for items they can work on.
Properly referencing the issue in the title ensures that Jira is correctly updated with code review comments and commits. -->
# Description
Please provide a short description of the changes you're making with this pull request.
# Solution
Please provide a short description of the approach taken to implement your solution.
# Tests
Please describe the tests you've developed or run to confirm this patch implements the feature or solves the problem.
# Checklist
Please review the following and check all that apply:
- [ ] I have reviewed the guidelines for [How to Contribute](https://wiki.apache.org/solr/HowToContribute) and my code conforms to the standards described there to the best of my ability.
- [ ] I have created a Jira issue and added the issue ID to my pull request title.
- [ ] I have given Solr maintainers [access](https://help.github.com/en/articles/allowing-changes-to-a-pull-request-branch-created-from-a-fork) to contribute to my PR branch. (optional but recommended)
- [ ] I have developed this patch against the `main` branch.
- [ ] I have run `./gradlew check`.
- [ ] I have added tests for my changes.
- [ ] I have added documentation for the [Reference Guide](https://github.com/apache/solr/tree/main/solr/solr-ref-guide)
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 commented on pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 commented on PR #1431:
URL: https://github.com/apache/solr/pull/1431#issuecomment-1461603109
> Wouldn't you cherry-pick it to `branch_9x`?
Ah yes - done!
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 commented on pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 commented on PR #1431:
URL: https://github.com/apache/solr/pull/1431#issuecomment-1460649318
Thanks @cpoerschke and @mkhludnev for your help with this!
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 commented on a diff in pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 commented on code in PR #1431:
URL: https://github.com/apache/solr/pull/1431#discussion_r1128286133
##########
solr/core/src/java/org/apache/solr/search/ReRankQParserPlugin.java:
##########
@@ -68,29 +71,51 @@ public Query parse() throws SyntaxError {
Query reRankQuery = reRankParser.parse();
int reRankDocs = localParams.getInt(RERANK_DOCS, RERANK_DOCS_DEFAULT);
- reRankDocs = Math.max(1, reRankDocs); //
+ reRankDocs = Math.max(1, reRankDocs);
double reRankWeight = localParams.getDouble(RERANK_WEIGHT, RERANK_WEIGHT_DEFAULT);
- return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight);
+ ReRankOperator reRankOperator =
+ ReRankOperator.get(localParams.get(RERANK_OPERATOR, RERANK_OPERATOR_DEFAULT));
+
+ return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight, reRankOperator);
}
}
private static final class ReRankQueryRescorer extends QueryRescorer {
- final double reRankWeight;
+ final BiFloatFunction scoreCombiner;
+
+ @FunctionalInterface
+ interface BiFloatFunction {
+ float func(float a, float b);
+ }
- public ReRankQueryRescorer(Query reRankQuery, double reRankWeight) {
+ public ReRankQueryRescorer(
+ Query reRankQuery, double reRankWeight, ReRankOperator reRankOperator) {
super(reRankQuery);
- this.reRankWeight = reRankWeight;
+ switch (reRankOperator) {
+ case ADD:
+ scoreCombiner = (score, second) -> (float) (score + reRankWeight * second);
+ break;
+ case MULTIPLY:
+ scoreCombiner = (score, second) -> (float) (score * reRankWeight * second);
+ break;
+ case REPLACE:
+ scoreCombiner = (score, second) -> (float) (reRankWeight * second);
Review Comment:
(Just FYI I'm away with work at the mo - am intending to take a look at this and probably merge it tomorrow eve.)
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 commented on a diff in pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 commented on code in PR #1431:
URL: https://github.com/apache/solr/pull/1431#discussion_r1129845797
##########
solr/core/src/java/org/apache/solr/search/ReRankQParserPlugin.java:
##########
@@ -68,29 +71,51 @@ public Query parse() throws SyntaxError {
Query reRankQuery = reRankParser.parse();
int reRankDocs = localParams.getInt(RERANK_DOCS, RERANK_DOCS_DEFAULT);
- reRankDocs = Math.max(1, reRankDocs); //
+ reRankDocs = Math.max(1, reRankDocs);
double reRankWeight = localParams.getDouble(RERANK_WEIGHT, RERANK_WEIGHT_DEFAULT);
- return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight);
+ ReRankOperator reRankOperator =
+ ReRankOperator.get(localParams.get(RERANK_OPERATOR, RERANK_OPERATOR_DEFAULT));
+
+ return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight, reRankOperator);
}
}
private static final class ReRankQueryRescorer extends QueryRescorer {
- final double reRankWeight;
+ final BiFloatFunction scoreCombiner;
+
+ @FunctionalInterface
+ interface BiFloatFunction {
+ float func(float a, float b);
+ }
- public ReRankQueryRescorer(Query reRankQuery, double reRankWeight) {
+ public ReRankQueryRescorer(
+ Query reRankQuery, double reRankWeight, ReRankOperator reRankOperator) {
super(reRankQuery);
- this.reRankWeight = reRankWeight;
+ switch (reRankOperator) {
+ case ADD:
+ scoreCombiner = (score, second) -> (float) (score + reRankWeight * second);
+ break;
+ case MULTIPLY:
+ scoreCombiner = (score, second) -> (float) (score * reRankWeight * second);
+ break;
+ case REPLACE:
+ scoreCombiner = (score, second) -> (float) (reRankWeight * second);
Review Comment:
Just tried a negative `reRankWeight`, and saw no surprises: it leads to negative final scores, with more negative scores pushing items down the result set.
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] cpoerschke commented on a diff in pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "cpoerschke (via GitHub)" <gi...@apache.org>.
cpoerschke commented on code in PR #1431:
URL: https://github.com/apache/solr/pull/1431#discussion_r1126644129
##########
solr/core/src/java/org/apache/solr/search/ReRankQParserPlugin.java:
##########
@@ -68,29 +71,51 @@ public Query parse() throws SyntaxError {
Query reRankQuery = reRankParser.parse();
int reRankDocs = localParams.getInt(RERANK_DOCS, RERANK_DOCS_DEFAULT);
- reRankDocs = Math.max(1, reRankDocs); //
+ reRankDocs = Math.max(1, reRankDocs);
double reRankWeight = localParams.getDouble(RERANK_WEIGHT, RERANK_WEIGHT_DEFAULT);
- return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight);
+ ReRankOperator reRankOperator =
+ ReRankOperator.get(localParams.get(RERANK_OPERATOR, RERANK_OPERATOR_DEFAULT));
+
+ return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight, reRankOperator);
}
}
private static final class ReRankQueryRescorer extends QueryRescorer {
- final double reRankWeight;
+ final BiFloatFunction scoreCombiner;
+
+ @FunctionalInterface
+ interface BiFloatFunction {
+ float func(float a, float b);
+ }
- public ReRankQueryRescorer(Query reRankQuery, double reRankWeight) {
+ public ReRankQueryRescorer(
+ Query reRankQuery, double reRankWeight, ReRankOperator reRankOperator) {
super(reRankQuery);
- this.reRankWeight = reRankWeight;
+ switch (reRankOperator) {
+ case ADD:
+ scoreCombiner = (score, second) -> (float) (score + reRankWeight * second);
+ break;
+ case MULTIPLY:
+ scoreCombiner = (score, second) -> (float) (score * reRankWeight * second);
+ break;
+ case REPLACE:
+ scoreCombiner = (score, second) -> (float) (reRankWeight * second);
Review Comment:
Wondering if `reRankWeight` at all or `reRankWeight` other than `1` makes sense for the `REPLACE` case? I guess it would allow scaling of the re-rank score.
Taking the `scaling of the ... score` thought further
```
(originalWeight * score + reRankWeight * second)
```
might be another way to achieve 'replace' logic i.e. with `originalWeight==0` value.
I have no strong views on this, just thinking out aloud.
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 merged pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 merged PR #1431:
URL: https://github.com/apache/solr/pull/1431
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 commented on a diff in pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 commented on code in PR #1431:
URL: https://github.com/apache/solr/pull/1431#discussion_r1128286133
##########
solr/core/src/java/org/apache/solr/search/ReRankQParserPlugin.java:
##########
@@ -68,29 +71,51 @@ public Query parse() throws SyntaxError {
Query reRankQuery = reRankParser.parse();
int reRankDocs = localParams.getInt(RERANK_DOCS, RERANK_DOCS_DEFAULT);
- reRankDocs = Math.max(1, reRankDocs); //
+ reRankDocs = Math.max(1, reRankDocs);
double reRankWeight = localParams.getDouble(RERANK_WEIGHT, RERANK_WEIGHT_DEFAULT);
- return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight);
+ ReRankOperator reRankOperator =
+ ReRankOperator.get(localParams.get(RERANK_OPERATOR, RERANK_OPERATOR_DEFAULT));
+
+ return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight, reRankOperator);
}
}
private static final class ReRankQueryRescorer extends QueryRescorer {
- final double reRankWeight;
+ final BiFloatFunction scoreCombiner;
+
+ @FunctionalInterface
+ interface BiFloatFunction {
+ float func(float a, float b);
+ }
- public ReRankQueryRescorer(Query reRankQuery, double reRankWeight) {
+ public ReRankQueryRescorer(
+ Query reRankQuery, double reRankWeight, ReRankOperator reRankOperator) {
super(reRankQuery);
- this.reRankWeight = reRankWeight;
+ switch (reRankOperator) {
+ case ADD:
+ scoreCombiner = (score, second) -> (float) (score + reRankWeight * second);
+ break;
+ case MULTIPLY:
+ scoreCombiner = (score, second) -> (float) (score * reRankWeight * second);
+ break;
+ case REPLACE:
+ scoreCombiner = (score, second) -> (float) (reRankWeight * second);
Review Comment:
(Just FYI I'm away with work at the mo - am intending to take a look at this and probably merge it tomorrow eve.)
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] mkhludnev commented on pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "mkhludnev (via GitHub)" <gi...@apache.org>.
mkhludnev commented on PR #1431:
URL: https://github.com/apache/solr/pull/1431#issuecomment-1461336728
Sure. @andywebb1975 . You are welcome. Wouldn't you cherry-pick it to `branch_9x`?
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org
[GitHub] [solr] andywebb1975 commented on a diff in pull request #1431: SOLR-16643: Add reRankOperator=multiply/replace options to ReRank query parser
Posted by "andywebb1975 (via GitHub)" <gi...@apache.org>.
andywebb1975 commented on code in PR #1431:
URL: https://github.com/apache/solr/pull/1431#discussion_r1126991270
##########
solr/core/src/java/org/apache/solr/search/ReRankQParserPlugin.java:
##########
@@ -68,29 +71,51 @@ public Query parse() throws SyntaxError {
Query reRankQuery = reRankParser.parse();
int reRankDocs = localParams.getInt(RERANK_DOCS, RERANK_DOCS_DEFAULT);
- reRankDocs = Math.max(1, reRankDocs); //
+ reRankDocs = Math.max(1, reRankDocs);
double reRankWeight = localParams.getDouble(RERANK_WEIGHT, RERANK_WEIGHT_DEFAULT);
- return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight);
+ ReRankOperator reRankOperator =
+ ReRankOperator.get(localParams.get(RERANK_OPERATOR, RERANK_OPERATOR_DEFAULT));
+
+ return new ReRankQuery(reRankQuery, reRankDocs, reRankWeight, reRankOperator);
}
}
private static final class ReRankQueryRescorer extends QueryRescorer {
- final double reRankWeight;
+ final BiFloatFunction scoreCombiner;
+
+ @FunctionalInterface
+ interface BiFloatFunction {
+ float func(float a, float b);
+ }
- public ReRankQueryRescorer(Query reRankQuery, double reRankWeight) {
+ public ReRankQueryRescorer(
+ Query reRankQuery, double reRankWeight, ReRankOperator reRankOperator) {
super(reRankQuery);
- this.reRankWeight = reRankWeight;
+ switch (reRankOperator) {
+ case ADD:
+ scoreCombiner = (score, second) -> (float) (score + reRankWeight * second);
+ break;
+ case MULTIPLY:
+ scoreCombiner = (score, second) -> (float) (score * reRankWeight * second);
+ break;
+ case REPLACE:
+ scoreCombiner = (score, second) -> (float) (reRankWeight * second);
Review Comment:
Thanks Christine! Yes, as far as I can see using `reRankWeight` is redundant for both `multiply` and `replace` - it’ll just scale the scores of all the reranked docs without changing their positions relative to each other - but I figured it'd be less surprising to be consistent between the operators rather than ignoring `reRankWeight` for some of them. This has got me wondering about negative weights though - I'll check what happens with those...
Using an `originalWeight` option would work too, yes - though I think it'll be clearer to have the explicit `reRankOperator=replace` in the query syntax so I'd prefer to stick with that. We'd still need the `reRankOperator=multiply` option anyway and it was trivial to add the extra `replace` option once `multiply` was there!
--
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: issues-unsubscribe@solr.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org