pr open: https://
github.com/hwchase17/lang
chain/pull/609
… @kweyamba_ai is this good for your use case?
@langchain
-
LangChain PR 609: New Features for Agent Development
By
–
-
LangChain Releases Collaborative Notebooks Gallery Section
By
–
Collab NB section in docs – @samcwl We've released a lot of really cool Collab Notebooks (LangChain-AGI, Wolfram Alpha, Tool updates) but haven't really tracked them anywhere Until now! Big shoutout to @samcwl for idea and implementation https://
langchain.readthedocs.io/en/latest/gall
ery.html#misc-colab-notebooks
… -
LangChain Exposes Custom Prompts for LLMMath and SQLite
By
–
Custom prompts Previously, prompts for LLMMath and SQLite chain were fixed Now they are exposed as parameters, meaning you can customize them! LLMMath: https://
langchain.readthedocs.io/en/latest/modu
les/chains/examples/llm_math.html
…
SQLite: https://
langchain.readthedocs.io/en/latest/modu
les/chains/examples/sqlite.html
… h/t @nickscamara_ Francisco Ingham in the discord for the idea -
v0.0.62 Release: Documentation and Text Processing Improvements
By
–
v0.0.62 Doc fixes – @eltociear @rukmal_w Sasmitha Manathunga Spacy Text Splitter fix – @SmitShah_11
Add filters to @pinecone – @samcarf Collab NB section in docs – @samcwl Custom prompt for LLMMath and SQLite Chains – h/t @nickscamara_ Francisco Ingham -
JSON Output Format for AI Model Integration
By
–
for that, its pretty dependent on your use case, but i would recommend telling it to write output in json (and showing it some examples) and then parsing that json
-
LangChain OutputParsers: Formatting LLM Outputs and Error Handling
By
–
its under documented, but checkout things like our OutputParsers – https://
github.com/hwchase17/lang
chain/blob/3e55f1474e5ad13f90c233b7f85480e996765e71/langchain/prompts/base.py#L67
… they can parse an LLM output into a specific format. however, it is an open question about what to do if format is not expected… retry? -
Return Source Documents in VectorDBQA Chain
By
–
Return source documents – Francisco Ingham You can now return documents used to generate the final answer in the VectorDBQA chain. Useful for getting more clarity on what was used to construct the answer Docs: https://
langchain.readthedocs.io/en/latest/modu
les/chains/combine_docs_examples/vector_db_qa.html#return-source-documents
… -
Pinecone Wrapper Updates: Namespace Support and Similarity Scores
By
–
Updates to @pinecone wrapper – @SmitShah_11 iocuydi @iamraymondyuan – Namespace support
– from_index classmethod (for existing indices)
– method to return similarity scores along with documents Hopefully making it easier to use Pinecone indices! -
LangChain VectorDBQA Chain Types Integration Methods
By
–
Chain types There are many methods for combining documents (stuff, map reduce, map rereank, refine) It is now SUPER easy to choose your favorite one to use in VectorDBQA chains! Docs for QA: https://
langchain.readthedocs.io/en/latest/modu
les/chains/combine_docs_examples/vector_db_qa.html#chain-type
… Docs for QA with sources: https://
langchain.readthedocs.io/en/latest/modu
les/chains/combine_docs_examples/vector_db_qa_with_sources.html
… -
LangChain v0.0.61 Release: Vector DB and WolframAlpha Improvements
By
–
v0.0.61 WolframAlpha improvements – @nickscamara_ Doc improvements – Marc Green Making it easier to work with Vector DBs Improve @pinecone wrapper – @SmitShah_11 iocuydi @iamraymondyuan Return source documents – Francisco Ingham
Chain types