Releases: KevKibe/docindex
v0.9.0
v0.8.0
Description
get_reranker_method
optimized to run in constant_time, O(1).- Dependabot configuration.
What's Changed
- Create dependabot.yml by @KevKibe in #28
- fix: run tests on pull_request_target by @KevKibe in #34
- Fix test workflow by @KevKibe in #35
- Chore(dependencies): Fix Dependabot workflow by @KevKibe in #36
- chore(dependabot): add permissions field by @KevKibe in #37
- Chore(dependabot): Fix dependabot test workflow by @KevKibe in #38
- Update: Optimize get reranker method to run in constant time complexity by @KevKibe in #39
- Update package version by @KevKibe in #40
Full Changelog: v0.7.0...v0.8.0
v0.7.0
New Features
Introducing intergration with reranking models from AnswerDotAI's rerankers inside the retrieve_and_generate step.
What's Changed
- Refactor: Move
create_index
,delete_index
,doc_model
,config
sub-modules into utils module. by @KevKibe in #22 - Add reranking step to
retrieve_and_generate
method in the Indexer classes by @KevKibe in #23 - chore(README): add rerank_model parameter to example by @KevKibe in #24
Full Changelog: v0.6.0...v0.7.0
v0.6.0
New Features
- RAG Retriever Integration: You can now set up a RAG (Retrieval Augmented Generation) retriever with sources using your vectorstore for efficient information retrieval.
- Cohere LLM Provider Support: Added support for Cohere as a language model provider, expanding your options for natural language processing tasks.
What's Changed
- adds retrieval logic by @koleshjr in #17
- Add Support for embedding, vectorstore initialization and rag retrieval using cohere. by @KevKibe in #18
- Update README.md by @KevKibe in #19
- Update setup.cfg by @KevKibe in #20
New Contributors
Full Changelog: v0.5.0...v0.6.0
v0.5.0
New Feature:
- HTML, Markdown, and DOCX document indexing is now supported. Contributed by @KevKibe. For more details, see #13.
Bug Fixes:
- None
Known Issues:
- No known issues at the moment.
Contributors:
Kevin Kibe ([email protected])
For more information, visit the project's GitHub repository: https://github.com/KevKibe/docindex
What's Changed
- Add new Diagram by @KevKibe in #11
- Chore: Add tests for
_google.docindex
and_openai.docindex
modules by @KevKibe in #12 - Feature: Support HTML, Markdown, and DOCX Document Indexing by @KevKibe in #13
- Update README.md by @KevKibe in #14
Full Changelog: v0.4.0...v0.5.0
v0.4.0
v0.3.0
New Features:
- Initialize a vectorstore object from the index created with the embedded documents inside.
Bug Fixes:
-None
Known Issues:
- No known issues at the moment.
Contributors:
Kevin Kibe ([email protected])
For more information, visit the project's GitHub repository: https://github.com/KevKibe/docindex
What's Changed
- Feat: Initialize vectorestore with
initialize_vectorstore
method from the index created by @KevKibe in #9
Full Changelog: v0.2.0...v0.3.0
v0.2.0
v0.1.0
New Features:
- Support for Google Embedding Service,
GoogleGenerativeAIEmbeddings
. - Support for indexing on the Command Line Interface(CLI)
Bug Fixes:
None
Known Issues:
No known issues at the moment.
Contributors:
Kevin Kibe ([email protected])
For more information, visit the project's GitHub repository: https://github.com/KevKibe/docindex
What's Changed
- Update: README.md Usage instructions, cli args for indexing by @KevKibe in #3
- Update: Support for Google Embedding Service,
GoogleGenerativeAIEmbeddings
by @KevKibe in #4 - Update: Support for Indexing using Google Generative AI Embeddings and README, DOCS Update by @KevKibe in #5
Full Changelog: v0.0.1...v0.1.0
v0.0.1
Release Notes - Version 0.0.1
- Initial release of the docindex package.
Features:
- Provides fast indexing of multiple PDF documents on the web or local directory and their metadata on Pinecone.
Bug Fixes:
- None
Known Issues:
- No known issues at the moment.
Contributors:
- Kevin Kibe ([email protected])
For more information, visit the project's GitHub repository: https://github.com/KevKibe/docindex