Langchain Conversational agent

[mrddter]

@dosu-bot

how could create a LangChain Conversational Agent with this requirements >

use only JavaScript
don't use OpenAI and OpenAIAgent (but if possible HF)
conversational chat able
memory remembering
give possibility to upload TXT to be used as a source of information
main purpose: the scope is its function as a Conversational Agent, with a focus on the history of exchanged messages, coupled with the use of data loaded from TXT files as a input source of info

[dosubot[bot]]

🤖

Based on the requirements you've provided, you can create a LangChain Conversational Agent using JavaScript, with memory remembering and conversational chat ability. However, the current context does not provide information on how to upload TXT files as a source of information.

Here's a basic example of how you can create a LangChain Conversational Agent:

import { BufferMemory } from "langchain/memory";
import { UpstashRedisChatMessageHistory } from "langchain/stores/message/upstash_redis";
import { HuggingFaceModel } from "langchain/chat_models/huggingface"; // This is hypothetical, as the context does not provide information on Hugging Face integration
import { ConversationChain } from "langchain/chains";

const memory = new BufferMemory({
  chatHistory: new UpstashRedisChatMessageHistory({
    sessionId: new Date().toISOString(),
    sessionTTL: 300,
    config: {
      url: "https://ADD_YOURS_HERE.upstash.io",
      token: "********",
    },
  }),
});

const model = new HuggingFaceModel({
  modelName: "gpt-3.5-turbo",
  temperature: 0,
});

const chain = new ConversationChain({ llm: model, memory });

const res1 = await chain.call({ input: "Hi! I'm Jim." });
console.log({ res1 });

const res2 = await chain.call({ input: "What did I just say my name was?" });
console.log({ res2 });

This code creates a new BufferMemory instance for memory remembering and a new HuggingFaceModel instance for the language model. It then creates a ConversationChain with these instances, which can be used for conversational chat ability. The call method is used to send messages to the chatbot and receive responses.

To integrate a Hugging Face model with LangChain in JavaScript, you can use the HuggingFaceTransformersEmbeddings class from the langchain/embeddings/hf_transformers module. Here's how you can do it:

import { HuggingFaceTransformersEmbeddings } from "langchain/embeddings/hf_transformers";

const model = new HuggingFaceTransformersEmbeddings({
  modelName: "Xenova/all-MiniLM-L6-v2",
});

/* Embed queries */
const res = await model.embedQuery(
  "What would be a good company name for a company that makes colorful socks?"
);
console.log({ res });
/* Embed documents */
const documentRes = await model.embedDocuments(["Hello world", "Bye bye"]);
console.log({ documentRes });

In this example, the model is used to embed a query and two documents, and the results are logged to the console.

Unfortunately, the current context does not provide information on how to upload TXT files as a source of information. If you could provide more details about how you want to use the TXT files, I might be able to provide a more complete answer.

Sources

• Chat History not beeing retrieved / saved
• examples/src/memory/upstash_redis.ts
• examples/src/memory/vector_store.ts
• langchain/src/llms/tests/huggingface_hub.int.test.ts
• examples/src/models/embeddings/hf_transformers.ts

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[reemster123]

You are asking a whole project. Is there anything specific you have questions about? Are there any blockers you ran into while trying to achieve this? I did ran into some, and also found some solutions for some, so maybe i could help, but the question is to wide as for now.

[reemster123]

You could start looking at this tutorial: https://learn.deeplearning.ai/functions-tools-agents-langchain/lesson/1/introduction

it is in python though, but a lot is similar. I also tried to replicate it in nodejs. You will run into some blockers since not everyting is the same in python/nodejs but at least its a good start