Grouping and Learning To Rank (SOLR-8776)

My company is very interested in using Learning To Rank in our product search. The problem we face is that our product search groups its results and that does not work with LTR. https://issues.apache.org/jira/browse/SOLR-8776 Is there any traction to getting the SOLR-8776 patch into the main

Does Learning To Rank feature require SolrCloud?

Hi, Do following features require SolrCloud? Or do they work in master-slave mode just fine? 1. Learning to rank (LTR) 2. Distributed IDF

Learning To Rank with Group Queries

Hi, My company is looking at using the Learning to Rank. However, our main searches do grouping. There is an old Jira from 2016 about how these don't work together. https://issues.apache.org/jira/browse/SOLR-8776 It doesn't look like this has moved much since then. When will we be a

The Learning To Rank result is treated as an input to a function

Hi all: How to embed a linear formula calculation in the outer layer of Learning To Rank? We need to do a calculation on the score of Learning To Rank. For example，normal use of Learning To Rank looks like this，I'm going to get a score: rq={!ltr model=myModel reRankDocs=100} I wan

About Learning To Rank

Excuse me, how to embed a linear formula calculation in the outer layer of LTR? We need to do a calculation on the score of LTR Thank

Solr Learning to Rank - Trailing slash in path behavior differences, expected?

Hello everyone, I encountered some surprising behavior that got be stuck on Solr LTR for a good hour. I wanted to share it, and you can decide if its a bug (I suspect it's a bug) I wanted to list all the feature stores on my Solr 7.7.1 instance. So I visited GET http://localhost:8983/solr/tmdb/s

Re: Learning to Rank Feature creation in python

Hi Ashis, Short answer: No, i don't think it's possible. I'm considering as well extending solr to allow plugging in features from outside, but it will require time because at the moment the features can see only the current document processed, while to do that ideally you want to process in one

Learning to Rank Feature creation in python

Hi , Hope you are doing well!!! I am currently trying to implement Learn To Rank in SOLR 7.5. I have gone through the blogs and documentation page to follow the steps. However the current process looks like i need to write SOLR query in a json file to upload the same to Feature Store. But the pr

Re: Learning to rank

Hi, Which version of Solr are you using? And do you have the error log for your error? Regards, Edwin On Mon, 16 Jul 2018 at 21:20, Akshay Patil wrote: > Hi > > I am student. for my master thesis I am working on the Learning To rank. As > I did research on it. I found solution

Learning to rank

Hi I am student. for my master thesis I am working on the Learning To rank. As I did research on it. I found solution provided by the Bloomberg. But I would like to ask. With the example that you have provided It always shows the error of Bad Request. Do you have running example of it. So i can

Re: Learning to rank - Bad Request

Hi Akshay, did you run solr enabling learning to rank? ./bin/solr -e techproducts -Dsolr.ltr.enabled=true if you don't pass -Dsolr.ltr.enabled=true ltr will not be available. Cheers, Diego From: solr-user@lucene.apache.org At: 07/16/18 09:00:39To: solr-user@lucene.apache.org Subjec

Re: Learning to rank - Bad Request

Hi, I am using apache solr 7.4.0. I am trying to use learning to rank using the python script and related data provided by the lucene. which can be found at the Github <https://github.com/apache/lucene-solr/tree/releases/lucene-solr/6.4.0/solr/contrib/ltr/example> repository of the lucen

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? Shou

Re: Learning to Rank (LTR) with grouping

Thanks ilayaraja, I updated the PR today integrating your and Alan's comments. Now it works also in distributed mode. Please let me know what do you think :) Cheers Diego On Wed, May 2, 2018, 17:46 ilayaraja wrote: > Figured out that offset is used as part of the grouping patch which I > appli

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() + ((AbstractReRankQuery)query).getReRankDocs()

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 Benedett

Re: Learning to Rank (LTR) with grouping

Are you using SolrCloud or any distributed search ? If you are using just a single Solr instance, LTR should have no problem with pagination. The re-rank involves the top K and then you paginate. So if a document from the original score page 1 ends up in page 3, you will see it at page three. have

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

Re: Learning to Rank (LTR) with grouping

de, but unit tests in non distribute mode works. > > Hope this helps, > Diego > > From: solr-user@lucene.apache.org At: 04/15/18 03:37:54To: > solr-user@lucene.apache.org > Subject: Re: Learning to Rank (LTR) with grouping > > People sometimes fill in the Fix/Version field

Re: Learning to Rank (LTR) with grouping

I just updated the PR to upstream - I still have to fix some things in distribute mode, but unit tests in non distribute mode works. Hope this helps, Diego From: solr-user@lucene.apache.org At: 04/15/18 03:37:54To: solr-user@lucene.apache.org Subject: Re: Learning to Rank (LTR) with

Re: Learning to Rank (LTR) with grouping

xample. >From the little of my opinion I believe it is quite useful. The last activity is from 22/May/17 15:23 and no committer commented after that. I would assume that committers with interest or expertise on Learning To Rank or Grouping initially didn't have free time to evaluate the pa

Re: Learning to Rank (LTR) with grouping

On 4/17/2018 5:35 AM, Alessandro Benedetti wrote: Apache Lucene/Solr is a big project, is there anywhere in the official Apache Lucene/Solr website where each committer list the modules of interest/expertise ? No, there is no repository like that.  Each committer knows what their own expertise

Re: Learning to Rank (LTR) with grouping

Hi Erick, I have a curiosity/suggestion regarding how to speed up pending( or forgotten ) Jiras, is there a way to find out the most suitable committer(s) for the task and tag them ? Apache Lucene/Solr is a big project, is there anywhere in the official Apache Lucene/Solr website where each commit

Re: Learning to Rank (LTR) with grouping

People sometimes fill in the Fix/Version field when they're creating the JIRA, since anyone can open a JIRA it's hard to control. I took that out just now. Basically if the "Resolution" field doesn't indicate it's fixed, you should assume that it hasn't been addressed. Patches welcome. Best, Eri

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 Ilay, I am still on Solr 6.6.0 and did not patch the grouping fix. I implemented a temporary workaround solution to have 2 async request from the web application 1st with grouping 2nd without grouping and merged the results. This solution worked for my case as we were getting grouping results f

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: http://lucene.472066.n3.nabble.com/Solr-User-f4720

Re: Using Solr to build a product matcher, with learning to rank

, wrote: > Hello, > > I'm considering using Solr with learning to rank to build a product matcher. > For example, it should match the titles: > - Apple iPhone 6 16 Gb, > - iPhone 6 16 Gb, > - Smartphone IPhone 6 16 Gb, > - iPhone 6 black 16 Gb, > to the same internal

Using Solr to build a product matcher, with learning to rank

Hello, I'm considering using Solr with learning to rank to build a product matcher. For example, it should match the titles: - Apple iPhone 6 16 Gb, - iPhone 6 16 Gb, - Smartphone IPhone 6 16 Gb, - iPhone 6 black 16 Gb, to the same internal reference, an unique identifier. With Solr,

Re: Learning to rank - Bad Request

Hi Vincent, Can you please elaborate more when u say u posted 'posted name feature one by one' ? I am getting the same error when i am trying to upload features for any other collection other than techproducts. -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

Re: Using 'learning to rank' with user specific features

Hi, let me see if I got your problem : your "user specific" features are Query dependent features from Solr side. The value of this feature depends on a query component ( the user Id) and a document component( product Id) You can definitely use them. You can model this feature as a binary feature.

Using 'learning to rank' with user specific features

hi, I am going through learning to rank examples in Solr7. In the examples, the features are part of the searched document. Can I use solr's learning to rank system if my features are user specific? e.g., if searching for products, i want to rank some products higher if they have been us

Re: Learning to Rank (LTR) with grouping

wrote: > Roopa, did you try collapsing instead of grouping? it would work with ltr.. > > From: solr-user@lucene.apache.org At: 01/11/18 16:48:26To: > solr-user@lucene.apache.org > Subject: Re: Learning to Rank (LTR) with grouping > > Solution that I implemented currently is: > S

Re: Learning to Rank (LTR) with grouping

Roopa, did you try collapsing instead of grouping? it would work with ltr.. From: solr-user@lucene.apache.org At: 01/11/18 16:48:26To: solr-user@lucene.apache.org Subject: Re: Learning to Rank (LTR) with grouping Solution that I implemented currently is: Since we have a web application which

Re: Learning to Rank (LTR) with grouping

> >>> Hi Roopa, >>> >>> >>> >>> If you look at the diff: >>> >>> >>> >>> https://github.com/apache/lucene-solr/pull/162/files >>> >>> >>> >>> I didn't change much in So

Re: Learning to Rank (LTR) with grouping

cher, you can try to skip the >> file >> >>> when applying the patch and redo the changes after. >> >>> >> >>> Alternatively, the feature branch is available here: >> >>> >> >>> https://github.com/bloomberg/lucene-so

Re: Learning to Rank (LTR) with grouping

; >>> > >>> Alternatively, the feature branch is available here: > >>> > >>> https://github.com/bloomberg/lucene-solr/commits/master-solr-8776 > >>> > >>> you could try to merge with that or cheery-pick my changes. > >>>

Re: Learning to Rank (LTR) with grouping

applying the patch and redo the changes after. >>> >>> Alternatively, the feature branch is available here: >>> >>> https://github.com/bloomberg/lucene-solr/commits/master-solr-8776 >>> >>> you could try to merge with that or cheery-pick my

Re: Learning to Rank (LTR) with grouping

groups or also in reranking the >> documents inside each group? >> >> Cheers, >> Diego >> >> >> From: solr-user@lucene.apache.org At: 12/09/17 19:07:25To: >> solr-user@lucene.apache.org >> Subject: Re: Learning to Rank (LTR) with grouping

Re: Learning to Rank (LTR) with grouping

> > From: solr-user@lucene.apache.org At: 12/09/17 19:07:25To: > solr-user@lucene.apache.org > Subject: Re: Learning to Rank (LTR) with grouping > > Hi I tried to apply this JIRA SOLR-8776 as a patch as this feature is > critical. > > Here are the steps I took on my mac: > > On br

Re: Learning to Rank (LTR) with grouping

-user@lucene.apache.org Subject: Re: Learning to Rank (LTR) with grouping Hi I tried to apply this JIRA SOLR-8776 as a patch as this feature is critical. Here are the steps I took on my mac: On branch branch_6_5 Your branch is up-to-date with 'origin/branch_6_5' patch -p1 -i 162.patch --dry-run

Re: Learning to Rank (LTR) with grouping

Hi I tried to apply this JIRA SOLR-8776 as a patch as this feature is critical. Here are the steps I took on my mac: On branch branch_6_5 Your branch is up-to-date with 'origin/branch_6_5' patch -p1 -i 162.patch --dry-run I am getting Failures for certain Hunks Example: patching file solr/c

Re: Learning to Rank (LTR) with grouping

Hi Diego, Thank you, I will look into this and see how I could patch this. Thank you for your quick response, Roopa On Fri, Dec 8, 2017 at 5:44 PM, Diego Ceccarelli wrote: > Hi Roopa, > > LTR is implemented using RankQuery, and at the moment grouping doens't > support RankQuery. > I opened a j

Re: Learning to Rank (LTR) with grouping

Hi Roopa, LTR is implemented using RankQuery, and at the moment grouping doens't support RankQuery. I opened a jira item time ago (https://issues.apache.org/jira/browse/SOLR-8776) and I would be happy to receive feedback on that. You can find the code here https://github.com/apache/lucene-solr/pu

Learning to Rank (LTR) with grouping

Hi, I am using grouping and LTR together and the results are not getting re-rank as it does without grouping. I am passing &rq parameter. Does LTR work with grouping on? Solr version 6.5 Thank you, Roopa

From Zero to Learning to Rank in Apache Solr

Here's a tutorial I wrote that some of you all might find useful: https://github.com/airalcorn2/Solr-LTR. Feedback is welcome. Thanks, Michael A. Alcorn

Re: How to Efficiently Extract Learning to Rank Similarity Features From Solr?

i think this can be actually a good idea and I think that would require a new feature type implementation. Specifically I think you could leverage the existing data structures ( such TermVector) to calculate the matrix and then perform the calculations you need. Or maybe there is space for even a

How to Efficiently Extract Learning to Rank Similarity Features From Solr?

Hi, I'm trying to extract several similarity measures from Solr for use in a learning to rank model. Doing this mathematically involves taking the dot product of several different matrices, which is extremely fast for non-huge data sets (e.g., millions of documents and queries). Howeve

Re: Learning to rank - Bad Request

I posted name feature one by one and thus it works. -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

Re: Learning to rank - Bad Request

What missing steps did you fix to solve the issue? i am facing exactly the same as you had before -- Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html

Re: Learning to rank - Bad Request

Hi, please could you post the relevant config steps? I am able to get the techproducts example working from here: https://github.com/apache/lucene-solr/tree/master/solr/contrib/ltr/example but if I change to another index I get the error "No REST managed resource registered for path /schema/fea

Learning-to-Rank with Bees: question answer follow-up

Hi everyone, At my "Learning-to-Rank with Apache Solr and Bees" talk on Friday [1] there was one question that wasn't properly understood (by me) and so not fully answered in the room but later in individual conversation the question/answer became clearer. So here I just want

Re: Solr learning to rank features question

ffZones:${zone}))" > > > > > > > > > (Function query documentation here: > > > https://cwiki.apache.org/confluence/display/solr/Function+Queries) > > > > > > Let me know if it works. > > > > > > Cheers, > > > Diego &

Re: Solr learning to rank features question

gt; PostCutOffZones:${zone}))" > > > > > > (Function query documentation here: > > https://cwiki.apache.org/confluence/display/solr/Function+Queries) > > > > Let me know if it works. > > > > Cheers, > > Diego > > > > > > &

Re: Solr learning to rank features question

ery( > PostCutOffZones:${zone}))" > > > (Function query documentation here: > https://cwiki.apache.org/confluence/display/solr/Function+Queries) > > Let me know if it works. > > Cheers, > Diego > > > > > > > > On Tue, Aug 29, 2017 at 5:47 PM, Bria

Re: Solr learning to rank features question

omplex solr filter query that I am hoping to turn into a > feature for learning to rank, but I am not sure if it's possible. I would > like to OR two fq together for one of my features. I have a cutoff date and > I need to check one field if NOW is before the cutoff, and a different >

Solr learning to rank features question

Hello, I have a somewhat complex solr filter query that I am hoping to turn into a feature for learning to rank, but I am not sure if it's possible. I would like to OR two fq together for one of my features. I have a cutoff date and I need to check one field if NOW is before the cutoff,

Fwd: Solr learning to rank features question

Hello, I have a somewhat complex solr filter query that I am hoping to turn into a feature for learning to rank, but I am not sure if it's possible. I would like to OR two fq together for one of my features. I have a cutoff date and I need to check one field if NOW is before the cutoff,

Re: SOLR Learning to Rank Questions

ou can use besides a standard Learning to Rank global model though to handle targeted per query rankings (click modeling, intent), personalization (different models per cluster), extremely trendy topics (online learning, bandits), etc, but this should be able handle most cases decently without

Re: Per Text Field Similarity Measures for Learning to Rank

rg/confluence/display/solr/Other+Schema+Elements>. This is an interesting approach and seems like it could solve your problem. Hope that helps -Mike On Fri, Aug 4, 2017 at 10:18 AM, Michael Alcorn wrote: > Hi all, > > I recently prototyped a learning to rank system in Python that produce

Re:SOLR Learning to Rank Questions

At: 08/03/17 11:15:10 ​ Dear all, First of all, I would like to thank you guys for the amazing job with SOLR. In special, I highly appreciate the learning to rank plugin. It is a fantastic work. I have two ​ ​ two questions for the LTR people and I hope this mailing list is the right place for

Per Text Field Similarity Measures for Learning to Rank

Hi all, I recently prototyped a learning to rank system in Python that produced promising results, so I'm now looking into how to replicate that process in our Solr setup. For my Python implementation, I was using a number of features that were per field text comparisons, e.g.:

SOLR Learning to Rank Questions

​ Dear all, First of all, I would like to thank you guys for the amazing job with SOLR. In special, I highly appreciate the learning to rank plugin. It is a fantastic work. I have two ​ ​ two questions for the LTR people and I hope this mailing list is the right place for that. *1)​ ​This is a

Re: extract multi-features for one solr feature extractor in solr learning to rank

plugin, this will involve a modification in how currently the feature vector is generated. Cheers - --- Alessandro Benedetti Search Consultant, R&D Software Engineer, Director Sease Ltd. - www.sease.io -- View this message in context: http://lucene.472066.n3.nabble.com/ex

Re: extract multi-features for one solr feature extractor in solr learning to rank

Although those features share the same class, the differences are minor. In product deployment, this overhead may not be an issue. After feature selection, probably only a small number of features are useful. Another use case: use convolution neural network or LSTM to extract embedded feature ve

Re: extract multi-features for one solr feature extractor in solr learning to rank

tion of the Feature Transformer <http://lucene.apache.org/solr/6_5_0/solr-ltr/org/apache/solr/ltr/response/transform/LTRFeatureLoggerTransformerFactory.html> to return features that actually triggered in a sparse format <https://cwiki.apache.org/confluence/display/solr/Learning+To+Rank#LearningT

extract multi-features for one solr feature extractor in solr learning to rank

Hi, I found that solr learning-to-rank (LTR) supports only ONE feature for a given feature extractor. See interface: https://github.com/apache/lucene-solr/blob/master/solr/contrib/ltr/src/java/org/apache/solr/ltr/feature/Feature.java Line (281, 282) (in FeatureScorer) @Override public

Re: Learning to rank - Bad Request

- Original Message - From: solr-user@lucene.apache.org To: solr-user@lucene.apache.org At: 03/06/17 13:22:40 Hi all, I've been trying to get learning to rank working on our own search index. Following the LTR-readme (https://github.com/bloomberg/lucene-solr/blob/master-ltr/solr/contrib/l

Re: Learning to rank - Bad Request

Hi all, I've been trying to get learning to rank working on our own search index. Following the LTR-readme (https://github.com/bloomberg/lucene-solr/blob/master-ltr/solr/contrib/ltr/example/README.md) I ran the example python script to train and upload the model, but I already get an error dur

Learning to rank - Bad Request

Hi all, I've been trying to get learning to rank working on our own search index. Following the LTR-readme (https://github.com/bloomberg/lucene-solr/blob/master-ltr/solr/contrib/ltr/example/README.md) I ran the example python script to train and upload the model, but I already get an

Re: How to train the model using user clicks when use ltr(learning to rank) module?

Hi, NLP4L[1] has not only Learning-to-Rank module but also a module which calculates click model and converts it into pointwise annotation data. NLP4L has a comprehensive manual[2], but you may want to read "Click Log Analysis" section[3] first to see if it suits your requirements.

Re: How to train the model using user clicks when use ltr(learning to rank) module?

ected. I may state the obvious, but if you can, try to collect as much signals type as possible. Cheers -- View this message in context: http://lucene.472066.n3.nabble.com/How-to-train-the-model-using-user-clicks-when-use-ltr-learning-to-rank-module-tp4312462p4313688.html Sent from the Solr - User mailing list archive at Nabble.com.

Re: How to train the model using user clicks when use ltr(learning to rank) module?

ah. very nice Diego. Thanks. On 1/6/2017 1:52 PM, Diego Ceccarelli (BLOOMBERG/ LONDON) wrote: Hi Jeffery, I submitted a patch to the README of the learning to rank example folder, trying to explain better how to produce a training set given a log with interaction data. Patch is available here

Re: How to train the model using user clicks when use ltr(learning to rank) module?

Hi Jeffery, I submitted a patch to the README of the learning to rank example folder, trying to explain better how to produce a training set given a log with interaction data. Patch is available here: https://issues.apache.org/jira/browse/SOLR-9929 And you can see the new version of the

Re: How to train the model using user clicks when use ltr(learning to rank) module?

k. Before the days of ML in Search. On 1/5/2017 3:33 PM, Jeffery Yuan wrote: Thanks , Will Martin. I checked the pdf it's great. but seems not very useful for my question: How to train the model using user clicks when use ltr(learning to rank) module. I know the concept after reading thes

Re: How to train the model using user clicks when use ltr(learning to rank) module?

Thanks , Will Martin. I checked the pdf it's great. but seems not very useful for my question: How to train the model using user clicks when use ltr(learning to rank) module. I know the concept after reading these papers. But still not sure how to code them. -- View this message in co

Re: How to train the model using user clicks when use ltr(learning to rank) module?

ps://static.aminer.org/pdf/PDF/000/472/865/optimizing_search_engines_using_clickthrough_data.pdf > http://www.cs.cornell.edu/people/tj/publications/joachims_etal_05a.pdf > http://alexbenedetti.blogspot.com/2016/07/solr-is-learning-to-rank-better-part-1.html > > But still have no clue how

How to train the model using user clicks when use ltr(learning to rank) module?

-is-learning-to-rank-better-part-1.html But still have no clue how to translate the partial pairwise feedback to the importance or relevance of that doc. >From a user's perspective, the steps such as setup the feature and model in Solr is simple, but collecting the feedback data and trai