v0.0.51 Unify return types for intermediate steps – h/t @johnjnay Add sports & knowledge graph results to SerpAPI wrapper – @brunotorious Document improvements: @eltociear @samcwl Making tools editable – h/t @brunotorious Memory support for chains with multi inputs
@langchain
-
Making Tools Editable: Customize Tool Descriptions for Better Agent Performance
By
–
Making tools editable When loading tools, they come with default descriptions. Sometimes, as @brunotorious pointed out, you may want to edit their descriptions in order to get them to work better with agents Now you can: https://
langchain.readthedocs.io/en/latest/exam
ples/agents/custom_tools.html#Modify-existing-tools
… -
Customizable AI Response Prefix in Conversational Memory
By
–
More customizability for conversational memory A small change, but allows for specification of the prefix to use for for the AI response Was previously hardcoded to "AI", but can now be set to whatever Idea came from the discord! https://
langchain.readthedocs.io/en/latest/exam
ples/memory/conversational_customization.html
… -
New Memory Type Combining Summary and Context Window
By
–
New memory type Previously, we had one memory type that created a summary, another that kept a window of recent lines of conversation. jaloo555 added a new type of memory which does both! It decides the window based on token length. https://langchain.readthedocs.io/en/latest/examples/memory/conversational_memory.html#ConversationSummaryBufferMemory
-
Weaviate Vectorstore: New add_texts Method Implementation
By
–
add_texts for @weaviate_io The @weaviate_io vectorstore implementation was previously missing an `add_texts` method @CShorten30 added one! This allows you to easily insert embeddings (eg, of examples for the example selector)
-
Map-Reduce and Refine Chains Now Return Intermediate Steps
By
–
Return intermediate steps for map-reduce & refine chains @johnjnay had the great idea to allow for returning of intermediate steps of these chains to the user This makes it easy to inspect things like Q&A over each document, before combining https://
langchain.readthedocs.io/en/latest/exam
ples/data_augmented_generation/question_answering.html#The-map_reduce-Chain
… -
Open Source Chatbot Creation Example Released
By
–
There's been lots of chatbots announced recently Check out for an open source example of how to create one! @brunotorious
-
Improved Vectorstore Creation from Multiple Documents
By
–
Easier support for creating vectorstores from many docs A common workflow is to embed many documents into a vectorstore @GradientAscent8 highlighted the need for improved functionality/documentation around this, and now it exists! https://
langchain.readthedocs.io/en/latest/exam
ples/data_augmented_generation/embeddings.html#From-Documents
… -
Multi-input Tools with Agents Documentation Improved
By
–
Multi-input tools with agents It wasn't super clear how to use tools that require multiple inputs with agents bryan on discord called this out, and as a result there is improved documentation! https://
langchain.readthedocs.io/en/latest/exam
ples/agents/multi_input_tool.html
… -
LangChain v0.0.47: Community Updates and Agent Improvements
By
–
v0.0.47 – lots of community initiated changes! Documentation Updates – @GradientAscent8 Easier support for creating vectorstores from many docs – s/o @GradientAscent8 Multi-input tools with agents Better parsing of agent output explanations: