Skip to content

[REQUEST]: Additional detail on how to get chunks from semantic_text #772

@kderusso

Description

@kderusso

Description

Semantic text has replaced chunks with highlighting. More details can be found in this blog post. This is leading to some confusion with internal and external users who were directly retrieving the chunks from the field. Now they have to use highlighting to see the chunks. This has caused a lot of questions.

It would be great to have in the docs a tutorial that explains how:

  • the text "chunks" are actually stored as start/end offset values
  • embeddings and offsets can be "viewed" with the field _inference_fields
  • maybe saying you can examine the text chunks by setting a high value for Highlight, but I'm mixed on that
  • Just something to point to when the question comes up and we won't have to spread secret tribal knowledge around the field

Resources

Explained in this blog: https://www.elastic.co/search-labs/blog/semantic-text-ga

Which documentation set does this change impact?

Elastic On-Prem and Cloud (all)

Feature differences

The feature is the same in all deployments.

What release is this request related to?

8.18

Collaboration model

The documentation team

Point of contact.

Main contact: @Mikep86 @kderusso

Stakeholders: @jeffvestal

Metadata

Metadata

Assignees

Labels

Team:DeveloperIssues owned by the Developer Docs Team

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions