Uploaded image for project: 'Solr'
  1. Solr
  2. SOLR-10081

Web UI (Generate Training Data ) for LTR plugin

    Details

    • Type: New Feature
    • Status: Open
    • Priority: Major
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: None
    • Security Level: Public (Default Security Level. Issues are Public)
    • Labels:
      None

      Description

      This is a New Feature for LTR plugin. It will help you to generate the training data (HUMAN_JUDGEMENT) by using score between 0-4 and generate (Click_logs). Live demo http://ltr.qfdk.me

        Issue Links

          Activity

          Hide
          githubbot ASF GitHub Bot added a comment -

          GitHub user qfdk opened a pull request:

          https://github.com/apache/lucene-solr/pull/154

          SOLR-10081: Web UI (GenerateTrainingData ) for LTR plugin

          Hello ,

          This is a New Feature for LTR plugin. It will help you to generate the training data (HUMAN_JUDGEMENT) by using score between 0-4. Live demo http://ltr.qfdk.me

          Thank you.

          You can merge this pull request into a Git repository by running:

          $ git pull https://github.com/qfdk/lucene-solr webui

          Alternatively you can review and apply these changes as the patch at:

          https://github.com/apache/lucene-solr/pull/154.patch

          To close this pull request, make a commit to your master/trunk branch
          with (at least) the following in the commit message:

          This closes #154


          commit c45bb40a8cca844540f2863202ac44aa147862d2
          Author: qfdk <qfdk@vip.qq.com>
          Date: 2017-01-31T21:26:39Z

          [Update] add GenTraingDataSolr webUI


          Show
          githubbot ASF GitHub Bot added a comment - GitHub user qfdk opened a pull request: https://github.com/apache/lucene-solr/pull/154 SOLR-10081 : Web UI (GenerateTrainingData ) for LTR plugin Hello , This is a New Feature for LTR plugin. It will help you to generate the training data (HUMAN_JUDGEMENT) by using score between 0-4. Live demo http://ltr.qfdk.me Thank you. You can merge this pull request into a Git repository by running: $ git pull https://github.com/qfdk/lucene-solr webui Alternatively you can review and apply these changes as the patch at: https://github.com/apache/lucene-solr/pull/154.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #154 commit c45bb40a8cca844540f2863202ac44aa147862d2 Author: qfdk <qfdk@vip.qq.com> Date: 2017-01-31T21:26:39Z [Update] add GenTraingDataSolr webUI
          Hide
          cpoerschke Christine Poerschke added a comment -

          Hello qfdk,

          Thank you for your interest in the Apache Solr Learning-to-Rank (LTR) plugin and thanks for opening this ticket.

          My initial thoughts would be that your GenTrainingDataSolr web UI is conceptually independent of any particular search engine and that as such the web UI would best be maintained and shared independently e.g. via a repo such as your GenTrainingDataSolr repo.

          In your repo seeing the Apache Solr vs. Apache Solr with LTR comparison in your second screenshot reminded me of the Quepid tool. If you haven't come across Quepid before then this blog post is great to get started.

          Regards,

          Christine


          PS: Upayavira and Alan Woodward - would you have any thoughts to share here perhaps e.g. based on your work with the Solr Admin UI and/or your experience with Quepid?

          Show
          cpoerschke Christine Poerschke added a comment - Hello qfdk , Thank you for your interest in the Apache Solr Learning-to-Rank (LTR) plugin and thanks for opening this ticket. My initial thoughts would be that your GenTrainingDataSolr web UI is conceptually independent of any particular search engine and that as such the web UI would best be maintained and shared independently e.g. via a repo such as your GenTrainingDataSolr repo. In your repo seeing the Apache Solr vs. Apache Solr with LTR comparison in your second screenshot reminded me of the Quepid tool. If you haven't come across Quepid before then this blog post is great to get started. Regards, Christine PS: Upayavira and Alan Woodward - would you have any thoughts to share here perhaps e.g. based on your work with the Solr Admin UI and/or your experience with Quepid?
          Hide
          qfdk qfdk added a comment -

          Christine Poerschke
          Thanks for your advice I think if we could developer an interface for ltr plugin is a good idea.

          exp:

          • if there is "visit_number" column in a database(solr) we could use this filed as "click_logs" by adding a button or an input filed in web ui.
          • A "train" button to execute the python code or javascript pour upload the feature and model. Then users could understand how ltr work.

          training data(WEB UI) -> feature -> model -> ltr query

          Regards,
          LI

          Show
          qfdk qfdk added a comment - Christine Poerschke Thanks for your advice I think if we could developer an interface for ltr plugin is a good idea. exp: if there is "visit_number" column in a database(solr) we could use this filed as "click_logs" by adding a button or an input filed in web ui. A "train" button to execute the python code or javascript pour upload the feature and model. Then users could understand how ltr work. training data(WEB UI) -> feature -> model -> ltr query Regards, LI

            People

            • Assignee:
              Unassigned
              Reporter:
              qfdk qfdk
            • Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

              • Created:
                Updated:

                Development