The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Add orcid for jlpereira.
- Add citation.cff.
- Fix typo
- Clean up API
- Fix naming, config
- Fix: gnbhl reference
- Improve chunks in result (#26)
- Add references to results (#25)
- Add bigger texts for JSON
- Prepare configuration of new features for Kubernetes.
v0.0.9 - 2023-12-24 Sat
- Add Cohere to rerank resuts (#24)
v0.0.8 - 2023-12-20 Wed
- Add Qdrant for data initiation (#23)
- Add Qdrant for querying (#21)
- Add Cross-Embedding (#22)
v0.0.7 - 2023-12-13 Wed
- Summary using OpenAi chatGPT 3.5 Turbo API (#20).
v0.0.6 - 2023-11-30
- Use model with bigger vectors
v0.0.5 - 2023-11-29
- Example questions for the web GUI (#19)
v0.0.4 - 2023-11-27
- Flag to show or hide matched texts (#18)
- Update web GUI form with GET data (#17)
v0.0.3 - 2023-11-23
- Add steps needed for Kubernetes deployment (#16)
- Add web GUI (#15)
v0.0.2 - 2023-11-17
- Add max-results and score-threshold parameters to API (#14)
- Add parameters for maximum number of returns and score threshold (#13)
- Change limiting factor from main_taxa to main_class of an item (#12)
- Add better handling of initiation process (#11)
- Add text of matched chunk to the output (if possible) (#10)
- Add an option to limit BHL items intake by a list of taxa (#9)
v0.0.1 - 2023-11-15
- Create API endpoint (#8).
- Ask a question and get an answer (#7).
- Embed chunks and save to the database (#6).
- Break items into chunks (#5).
- Connect to
llmutilRESTful service (#4). - Create database reset with
pgvectorfor the data (#3). - Return list of item IDs (#2).
v0.0.0 - 2023-11-11
- General "plumbing": files structure, makefile etc. (#1)
This document follows changelog guidelines