Discource-C
/
discourse-ai
mirror of https://github.com/discourse/discourse-ai.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

FIX: Invalidate semantic search cache entries when hyde or embedding model changes (#202)

This commit is contained in:
Roman Rizzi 2023-09-05 18:39:39 -03:00 committed by GitHub
parent 3a00c38fc5
commit 4d854e9232
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 33 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
assets/javascripts/discourse/connectors/full-page-search-below-search-header/semantic-search.js
View File

@ -16,7 +16,7 @@ export default class extends Component {
@service appEvents;
@tracked searching = false;
@tracked searching = true;
@tracked collapsedResults = true;
@tracked results = [];
@ -57,7 +57,7 @@ export default class extends Component {
@bind
performHyDESearch() {
if (!this.searchTerm || !this.searchEnabled || this.searching) {
if (!this.searchTerm || !this.searchEnabled) {
return;
}

38
lib/modules/embeddings/semantic_search.rb
View File

@ -6,8 +6,11 @@ module DiscourseAi
def self.clear_cache_for(query)
digest = OpenSSL::Digest::SHA1.hexdigest(query)
Discourse.cache.delete("hyde-doc-#{digest}")
Discourse.cache.delete("hyde-doc-embedding-#{digest}")
hyde_key =
"semantic-search-#{digest}-#{SiteSetting.ai_embeddings_semantic_search_hyde_model}"
Discourse.cache.delete(hyde_key)
Discourse.cache.delete("#{hyde_key}-#{SiteSetting.ai_embeddings_model}")
end
def initialize(guardian)
@ -16,7 +19,14 @@ module DiscourseAi
def cached_query?(query)
digest = OpenSSL::Digest::SHA1.hexdigest(query)
Discourse.cache.read("hyde-doc-embedding-#{digest}").present?
embedding_key =
build_embedding_key(
digest,
SiteSetting.ai_embeddings_semantic_search_hyde_model,
SiteSetting.ai_embeddings_model,
)
Discourse.cache.read(embedding_key).present?
end
def search_for_topics(query, page = 1)
@ -29,11 +39,19 @@ module DiscourseAi
DiscourseAi::Embeddings::VectorRepresentations::Base.current_representation(strategy)
digest = OpenSSL::Digest::SHA1.hexdigest(query)
hyde_key = build_hyde_key(digest, SiteSetting.ai_embeddings_semantic_search_hyde_model)
embedding_key =
build_embedding_key(
digest,
SiteSetting.ai_embeddings_semantic_search_hyde_model,
SiteSetting.ai_embeddings_model,
)
hypothetical_post =
Discourse
.cache
.fetch("hyde-doc-#{digest}", expires_in: 1.week) do
.fetch(hyde_key, expires_in: 1.week) do
hyde_generator = DiscourseAi::Embeddings::HydeGenerators::Base.current_hyde_model.new
hyde_generator.hypothetical_post_from(query)
end
@ -41,9 +59,7 @@ module DiscourseAi
hypothetical_post_embedding =
Discourse
.cache
.fetch("hyde-doc-embedding-#{digest}", expires_in: 1.week) do
vector_rep.vector_from(hypothetical_post)
end
.fetch(embedding_key, expires_in: 1.week) { vector_rep.vector_from(hypothetical_post) }
candidate_topic_ids =
vector_rep.asymmetric_topics_similarity_search(
@ -63,6 +79,14 @@ module DiscourseAi
private
attr_reader :guardian
def build_hyde_key(digest, hyde_model)
"semantic-search-#{digest}-#{hyde_model}"
end
def build_embedding_key(digest, hyde_model, embedding_model)
"#{build_hyde_key(digest, hyde_model)}-#{embedding_model}"
end
end
end
end