discourse/discourse-ai
1
0
Fork 0
mirror of https://github.com/discourse/discourse-ai.git synced 2025-03-06 17:30:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

FIX: Embeddings backfill rake task was using old code (#1084)

Browse Source
This commit is contained in:
Roman Rizzi 2025-01-22 14:00:26 -03:00 committed by GitHub
parent e2e753d73c
commit faa8e6e873
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 11 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
lib/tasks/modules/embeddings/database.rake
View File

@ -1,24 +1,21 @@
# frozen_string_literal: true
desc "Backfill embeddings for all topics and posts"
task "ai:embeddings:backfill", %i[model concurrency] => [:environment] do |_, args|
task "ai:embeddings:backfill", %i[embedding_def_id concurrency] => [:environment] do |_, args|
public_categories = Category.where(read_restricted: false).pluck(:id)
if args[:model].present?
strategy = DiscourseAi::Embeddings::Strategies::Truncation.new
vector_rep =
DiscourseAi::Embeddings::VectorRepresentations::Base.find_representation(args[:model]).new(
strategy,
)
if args[:embedding_def_id].present?
vdef = EmbeddingDefinition.find(args[:embedding_def_id])
vector_rep = DiscourseAi::Embeddings::Vector.new(vdef)
else
vector_rep = DiscourseAi::Embeddings::VectorRepresentations::Base.current_representation
vector_rep = DiscourseAi::Embeddings::Vector.instance
end
table_name = DiscourseAi::Embeddings::Schema::TOPICS_TABLE
topics_table_name = DiscourseAi::Embeddings::Schema::TOPICS_TABLE
topics =
Topic
.joins("LEFT JOIN #{table_name} ON #{table_name}.topic_id = topics.id")
.where("#{table_name}.topic_id IS NULL")
.joins("LEFT JOIN #{topics_table_name} ON #{topics_table_name}.topic_id = topics.id")
.where("#{topics_table_name}.topic_id IS NULL")
.where("category_id IN (?)", public_categories)
.where(deleted_at: nil)
.order("topics.id DESC")
@ -29,11 +26,11 @@ task "ai:embeddings:backfill", %i[model concurrency] => [:environment] do |_, ar
end
end
table_name = vector_rep.post_table_name
posts_table_name = DiscourseAi::Embeddings::Schema::POSTS_TABLE
posts =
Post
.joins("LEFT JOIN #{table_name} ON #{table_name}.post_id = posts.id")
.where("#{table_name}.post_id IS NULL")
.joins("LEFT JOIN #{posts_table_name} ON #{posts_table_name}.post_id = posts.id")
.where("#{posts_table_name}.post_id IS NULL")
.where(deleted_at: nil)
.order("posts.id DESC")