langchainjs
langchainjs copied to clipboard
hwchase17
### Describe the issue When I enable streaming on the OpenAI model, it causes all answers to begin with a rephrased version of the question. #### Example: Question from user:...
After receiving this error from SerpAPI tool  I want to add an optional parameter to `BaseParameters` to include a fallback string message in...
Loading the following agent fails with "UnhandledPromiseRejectionWarning: Error: Unknown agent type": https://github.com/hwchase17/langchain-hub/tree/master/agents/zero-shot-react-conversation ``` const model = new ChatOpenAI({ temperature: 0}); const tools: Tool[] = [new SerpAPI(), new Calculator()]; const agent...
The docs could use a clear example of agent composition to make it obvious that agents can be used to coordinate actions with other agents doing intermediary work. To make...
Not really an issue more of inspiration for a weaviate vectorstore: ``` import { Embeddings } from "langchain/embeddings"; import { VectorStore } from "langchain/vectorstores"; import { IWeaviateClient } from "weaviate-ts-client";...
Based on OPENAI Cook book example, pass example_user and example_assisatant in message name is a way of few shot example, link is here [](https://github.com/openai/openai-cookbook/blob/main/examples/How_to_format_inputs_to_ChatGPT_models.ipynb). response = openai.ChatCompletion.create( model=MODEL, messages=[ {"role":...
The problem: sometimes, files in the repo might exceed recommended token size, meaning that vector databases like Pinecone won't be able to process such documents. In ideal world, I'd imagine...
I found that `MapReduceDocumentsChain` implements `StuffDocumentsChainInput`, but `MapReduceDocumentsChainInput` was not being used. So I updated it and fixed a typo.
Works with xata.io serverless database (with vector storage support) Feedback welcome! I tried to stay very similar to the supabase implementation. Tests not added as it would get messy due...
