This is in addition to Obsidian, Google Drive, Microsoft Word, PowerPoints, and more What should we add next that would enable your #chatyourdata submission?
@langchain
-
LangChain Introduces Async Support for LLMs Chains Agents
By
–
Async Support in LangChain It's been a long time coming, one of our most requested features, and it's finally here for select LLMs, chains, and agents. Blog Post:
-
LangChain Async APIs for LLMs Chains and Agents
By
–
Relevant Docs: Async API for LLMs: https://
langchain.readthedocs.io/en/latest/modu
les/llms/async_llm.html
… Async API for Chains: https://
langchain.readthedocs.io/en/latest/modu
les/chains/async_chain.html
… Async API for Agents: https://
langchain.readthedocs.io/en/latest/modu
les/chains/async_chain.html
… -
Adding Example Rows Improves SQL Database Chain Performance
By
–
SQLDatabase chain # of rows Research by Nitarshan Rajkumar, Raymond Li, Dzmitry Bahdanau shows that the more rows you add into the prompt, the better you can query SQL s/o to Jon Luo for adding Paper: https://
arxiv.org/abs/2204.00498 Example: https://
langchain.readthedocs.io/en/latest/modu
les/chains/examples/sqlite.html#adding-example-rows-from-each-table
… -
Google Drive and Obsidian Data Loaders for LangChain
By
–
Google Drive and Obsidian data loaders More data loading support! Helpful for the Chat-Your-Data challenge Google Drive: https://
langchain.readthedocs.io/en/latest/modu
les/document_loaders/examples/googledrive.html
… Obsidian: https://
langchain.readthedocs.io/en/latest/modu
les/document_loaders/examples/obsidian.html
… -
FAISS Vectorstore Saving and Loading Improvements
By
–
Saving/loading of FAISS vectorstore We heard a lot that saving/loading of FAISS vectorstore was hard – now its hopefully a bit easier s/o Shrey Joshi for fixing, and Matthew Bilyeu, Auro Tripathy, Amey narayan and xloem for the discussions Docs: https://
langchain.readthedocs.io/en/latest/modu
les/utils/combine_docs_examples/vectorstores.html#saving-and-loading
… -
Open Source Project Updates and Bug Fixes Released
By
–
And lots of misc bugs/docs improvements! s/o @rvorias
, Zack Witten, Gabriel Simmons, @kevin_w_huo
, Amos Ng, @jamescalam for flagging and/or fixing! -
LangChain 0.0.79 Release: New AI Features and Integrations
By
–
New release (0.0.79) with new features! @huggingface inference endpoint integration SQLDatabase chain # of rows Google Drive and Obsidian data loaders Saving/loading of FAISS vectorstore And also lots of small quality of life bug fixes/doc improvements
-
Hugging Face Inference Endpoint Integration in LangChain
By
–
@huggingface inference endpoint integration If you've deployed your Hugging Face model to an inference endpoint, you can use it easily in LangChain! s/o @EnoReyes for adding Inference Endpoint Docs: https://
huggingface.co/inference-endp
oints
… Code: https://
github.com/hwchase17/lang
chain/blob/master/langchain/llms/huggingface_endpoint.py
… -
LangChain Methods for Handling Context Window Limits
By
–
check out documentation here: https://
langchain.readthedocs.io/en/latest/use_
cases/question_answering.html
… theres a few different methods for getting around the context window limits!