Skip to content

Commit

Permalink
Reopen mtern descriptions PR for editorial. (#648)
Browse files Browse the repository at this point in the history
* Reopen mtern descriptions PR for editorial.

Signed-off-by: Archer <[email protected]>

* Apply suggestions from code review

Co-authored-by: Nathan Bower <[email protected]>
Signed-off-by: Naarcha-AWS <[email protected]>

* Remove spaces

Signed-off-by: Archer <[email protected]>

---------

Signed-off-by: Archer <[email protected]>
Signed-off-by: Naarcha-AWS <[email protected]>
Co-authored-by: Nathan Bower <[email protected]>
  • Loading branch information
Naarcha-AWS and natebower authored Oct 25, 2024
1 parent 0605aec commit 931a61a
Show file tree
Hide file tree
Showing 2 changed files with 9 additions and 9 deletions.
2 changes: 1 addition & 1 deletion spec/schemas/_core.mtermvectors.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ components:
_index:
$ref: '_common.yaml#/components/schemas/IndexName'
doc:
description: An artificial document for which you want to retrieve term vectors.
description: An artificial document for which you want to retrieve term vectors.
type: object
fields:
$ref: '_common.yaml#/components/schemas/Fields'
Expand Down
16 changes: 8 additions & 8 deletions spec/schemas/_core.rank_eval.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ components:
template_id:
$ref: '_common.yaml#/components/schemas/Id'
params:
description: The search template parameters.
description: The search template parameters.
type: object
additionalProperties:
type: object
Expand Down Expand Up @@ -70,7 +70,7 @@ components:
- type: object
properties:
ignore_unlabeled:
description: Controls how unlabeled documents in the search results are counted. When `true`, unlabeled documents are ignored and do not count as relevant or irrelevant. When `false`, unlabeled documents are treated as irrelevant.
description: Controls how unlabeled documents in the search results are counted. When `true`, unlabeled documents are ignored and are not treated as relevant or irrelevant. When `false`, unlabeled documents are treated as irrelevant.
type: boolean
RankEvalMetricRatingThreshold:
allOf:
Expand All @@ -84,7 +84,7 @@ components:
type: object
properties:
k:
description: Sets the maximum number of documents retrieved per query. This value replaces the usual `size` parameter in the query.
description: Sets the maximum number of documents retrieved per query. This value replaces the `size` parameter in the query.
type: number
RankEvalMetricRecall:
allOf:
Expand All @@ -100,15 +100,15 @@ components:
- type: object
properties:
normalize:
description: When `true`, this metric calculates the Normalized DCG (https://en.wikipedia.org/wiki/Discounted_cumulative_gain#Normalized_DCG).
description: When `true`, calculates the [normalized discounted cumulative gain (nDCG)](https://en.wikipedia.org/wiki/Discounted_cumulative_gain#Normalized_DCG).
type: boolean
RankEvalMetricExpectedReciprocalRank:
allOf:
- $ref: '#/components/schemas/RankEvalMetricBase'
- type: object
properties:
maximum_relevance:
description: The highest relevance grade used in the user-supplied relevance judgments.
description: The highest relevance grade used in user-supplied relevance judgments.
type: number
required:
- maximum_relevance
Expand All @@ -119,17 +119,17 @@ components:
description: The `metric_score`, found in the `metric_details` section, shows the contribution of this query to the global quality metric score.
type: number
unrated_docs:
description: The `unrated_docs` section contains an `_index` and `_id` entry for each document that didn't have a ratings value. This can be used to ask the user to supply ratings for these documents.
description: The `unrated_docs` section contains an `_index` and `_id` entry for each document that didn't have a `ratings` value. This can be used to ask the user to supply ratings for these documents.
type: array
items:
$ref: '#/components/schemas/UnratedDocument'
hits:
description: The `hits` section shows a grouping of the search results with their supplied ratings
description: The `hits` section provides a grouping of the search results with their supplied ratings.
type: array
items:
$ref: '#/components/schemas/RankEvalHitItem'
metric_details:
description: The `metric_details` section gives additional information about the calculated quality metric, in other words, how many of the retrieved documents were relevant. The content varies for each metric but allows for better interpretation of the results.
description: The `metric_details` section provides additional information about the calculated quality metric indicating the number of relevant retrieved documents. The content varies for each metric but allows for better interpretation of the results.
type: object
additionalProperties:
type: object
Expand Down

0 comments on commit 931a61a

Please sign in to comment.