discourse-ai/spec/lib/modules/embeddings
Sam dcafc8032f
FIX: improve embedding generation (#452)
1. on failure we were queuing a job to generate embeddings, it had the wrong params. This is both fixed and covered in a test.
2. backfill embedding in the order of bumped_at, so newest content is embedded first, cover with a test
3. add a safeguard for hidden site setting that only allows batches of 50k in an embedding job run

Previously old embeddings were updated in a random order, this changes it so we update in a consistent order
2024-01-31 10:38:47 -03:00
..
jobs FEATURE: Move the default embeddings model to bge-large-en (#417) 2024-01-11 14:16:25 -03:00
strategies FEATURE: HyDE-powered semantic search. (#136) 2023-09-05 11:08:23 -03:00
vector_representations FEATURE: Per post embeddings (#387) 2023-12-29 12:28:45 -03:00
entry_point_spec.rb FEATURE: Setting to control per post embeddings (#439) 2024-01-23 22:09:27 -03:00
semantic_related_spec.rb FIX: improve embedding generation (#452) 2024-01-31 10:38:47 -03:00
semantic_search_spec.rb UX: Validations to LLM-backed features (except AI Bot) (#436) 2024-01-29 16:04:25 -03:00
semantic_topic_query_spec.rb DEV: Update linting (#423) 2024-01-13 00:28:06 +01:00