Feature stores are a pillar of traditional machine learning. But do they transfer to the LLM era? I think they do! Here's a blog diving into how and why you should use them: https://
blog.langchain.dev/feature-stores
-and-llms/
… Here's an example notebook for @langchain
: https://
python.langchain.com/en/latest/modu
les/prompts/prompt_templates/examples/connecting_to_a_feature_store.html
…
@hwchase17
-
Feature Stores Integration with Large Language Models
By
–
-
Impressive Creative AI Application with Sequential Chain Architecture
By
–
This is a crazy impressive AND creative application
— Harrison Chase (@hwchase17) 8 mai 2023
Uses a SequentialChain of **10** other chains(!!) https://t.co/5pLF89yhVBThis is a crazy impressive AND creative application Uses a SequentialChain of **10** other chains(!!)
-
Putting AI Agents into Production: Lessons and Best Practices
By
–
There's been lots of exploration with Agents. But very few of them have actually made it into production. Join the next @langchain webinar with @sjwhitmore @DivGarg9 and @devstein64 to understand what it looks like to put an agent in production
-
LangChain Webinars Now on YouTube with QA Demo
By
–
We just put all the @langchain webinars on Youtube! https://
youtube.com/channel/UCC-ly
oTfSrcJzA1ab3APAgw
… The best part is, this now means we can use LangChain (in 12 lines of code) to do question/answering over the LangChain webinars How circular Gist: https://
gist.github.com/hwchase17/aa78
ad6fcb43d84665bdf2cf03afbc8e
… -
Agent Architecture: Batch Tool Invocations vs Sequential Observation
By
–
Like outputs a list of tool invocations? Rather than an agent which outputs one, then observes, then another, then observes, etc
-
HumanInputLLM: LLM Wrapper for Agent Debugging
By
–
HumanInputLLM A LLM wrapper that: 1. prints out the full prompt
2. asks the user for input to use as the response Useful for stepping through and debugging complex agents/chains in a controlled manner! Thanks to Myeongseop Kim for adding Docs: https://
python.langchain.com/en/latest/modu
les/models/llms/examples/human_input_llm.html
… -
Kor: Deep Dive into Entity Parsing Tool
By
–
absolutely! id recomend checking out kor for entity parsing deep dive https://
github.com/eyurtsev/kor -
ChromaDB Self-Querying Retriever with LLM Integration
By
–
ChromaDB Self-Querying Retriever Last week we introduced the self-querying retriever Basic idea is to use an LLM to turn a user query into a "query" and a "filter" We now implemented to work with @trychroma
! Docs: https://
github.com/hwchase17/lang
chain/blob/master/docs/modules/indexes/retrievers/examples/chroma_self_query_retriever.ipynb
… -
How AI Agents Currently Select Tools via LLM Decisions
By
–
wdym by "beyond comparing description embeddings"? thats not how agents currently work. they currently work by asking the llm to pick a tool, very similar to this approach
-
LangChain Wikibase Agent Implementation and Database Examples
By
–
theres one for wikibase: https://
python.langchain.com/en/latest/use_
cases/agents/wikibase_agent.html
… got any other good example databases to implement one on?