Re: Re:LTR performance issues

Thanks, Deigo. I shall followup from the jira.. - --Ilay -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

LTR performance issues

LTR with grouping results in very high latency (3x) even while re-ranking 24 top groups. How is re-ranking implemented in Solr? Is it expected that it would result in 3x more query time. Need clarifications on: 1. How many top groups are actually re-ranked, is it exactly what we pass in

LTR performance issues

LTR with grouping results in very high latency (3x) even while re-ranking 24 top groups. How is re-ranking implemented in Solr? Is it expected that it would result in 3x more query time. Need clarifications on: 1. How many top groups are actually re-ranked, is it exactly what we pass in

Re: Learning to Rank (LTR) with grouping

Also, would like to understand what are the ways to optimize for performance at search time with LTR. Queries with terms (that fetch more results) lead to very high latency with re-rank query even for reRankDocs=24. Is there best practices to reduce the latency? Can fv cache help?

Re: Learning to Rank (LTR) with grouping

Figured out that offset is used as part of the grouping patch which I applied (SOLR-8776) : solr/core/src/java/org/apache/solr/handler/component/QueryComponent.java + if (query instanceof AbstractReRankQuery){ +topNGroups = cmd.getOffset() +

Re: Learning to Rank (LTR) with grouping

* "Top K shouldn't start from the "start" parameter, if it does, it is a bug. "*** 1. I clearly see that LTR do re-rank based on the start parameter. 2. When reRankDocs=24, pageSize=24, I still get the second page of results re-ranked by ltr plugin when I query with start=24. Alessandro

Re: Learning to Rank (LTR) with grouping

Between, I have applied the patch on top of solr 7.2.1 and it worked well for me though the Test Cases were failing, yet to see why. On another note, LTR with reRankDocs>page_size seems to create issue. For example, Say my page_size=24 and reRankDocs=48. For first query with start=0, it returns

Re: Learning to Rank (LTR) with grouping

Between, I have applied the patch on top of solr 7.2.1 and it worked well for me though the Test Cases were failing, yet to see why. On another note, LTR with reRankDocs>page_size seems to create issue. For example, Say my page_size=24 and reRankDocs=48. For first query with start=0, it returns

Support LTR RankQuery with Grouping

I am facing issue with LTR query not supported with grouping. I see the patch for this has been raised here https://issues.apache.org/jira/browse/SOLR-8776 Is it available in solr/master (7.2.2) now? Looks like this patch is not merged yet. - --Ilay -- Sent from:

Re: Learning to Rank (LTR) with grouping

Thanks Roopa. I was expecting that the issue has been fixed in solr 7.0 as per here https://issues.apache.org/jira/browse/SOLR-8776. Let me see why it is still not working on solr-ltr-7.2.1 - --Ilay -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

Re: Learning to Rank (LTR) with grouping

Hi Roopa & Deigo, I am facing same issue with grouping. Currently, am on Solr 7.2.1 but still see that grouping with LTR is not working. Did you apply it as patch or the latest solr version has the fix already? Ilay - --Ilay -- Sent from:

Re: LTR - OriginalScore query issue

Yes, I have tried that too: But it was throwing error while feature extraction: "Exception from createWeight for SolrFeature [name=originalLuceneScore, params={q={!dismax qf=tem_type_all^30.0 ..}${user_query}}] Failed to parse feature query. at

Re: LTR - OriginalScore query issue

I do have the features defined as below for field specific (title..) matching etc: features: [ { name: "productNewness", class: "org.apache.solr.ltr.feature.SolrFeature", params: { q: "{!func}recip( ms(NOW,launchdate_pl), 3.16e-11, 1, 1)" }, store: "myFeatureStoreDemo", }, { name:

LTR - OriginalScore query issue

solr/collection/select?fl=id,score,[features+store=myFeatureStore+efi.user_query='black shoes']=json=id:"13245336"=on When we fire this query during feature extraction, the originalScore feature gets the score of the "id" match but not the actual user query which is in this case 'black shoes'.

Re: Solr LTR plugin - Training

Also, when I want to add phrase match as a feature as below, it does not support it: { "store" : "tsf", "name" : "productTitleMatchGuestQuery", "class" : "org.apache.solr.ltr.feature.SolrFeature", "params" : { "q" : "{!dismax qf=p_title^1.0}${user_query}" } }, { "store" :

Re: LTR training

Thanks Diego for the pointers..will check. - --Ilay -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

Re: Solr LTR plugin - Training

Yes, that works. Thanks. - --Ilay -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

Solr autocomplete feature

Hi I want to change autocomplete implementation for our search. Current I have a suggest field whose definition in schema.xml is as below: field name=suggest type=edgytext indexed=true stored=true required=true omitNorms=false/ fieldType name=edgytext class=solr.TextField