Is there any option to use Azure open ai chat completion use your own data(data source) with langchain?

[nipun-threestops]

I have started using langchain with azure chat completion model hosted.I also needed to pass data sources to the chat completion model as in azure playground/ REST api . IS there any option to do so?

[dosubot[bot]]

🤖

Hello,

Yes, you can use your own data sources with the Azure OpenAI chat completion model in LangChain. You can pass tools or functions to the model. These tools or functions can be used to fetch data from your own data sources.

Here is an example of how you can do this:

const modelForFunctionCalling = new ChatOpenAI({
  modelName: "gpt-4",
  temperature: 0,
});

await modelForFunctionCalling.predictMessages(
  [new HumanMessage("What is the weather in New York?")],
  {
    functions: [
      {
        name: "get_current_weather",
        description: "Get the current weather in a given location",
        parameters: {
          type: "object",
          properties: {
            location: {
              type: "string",
              description: "The city and state, e.g. San Francisco, CA",
            },
            unit: { type: "string", enum: ["celsius", "fahrenheit"] },
          },
          required: ["location"],
        },
      },
    ],
    // You can set the `function_call` arg to force the model to use a function
    function_call: {
      name: "get_current_weather",
    },
  }
);

In this example, a function named get_current_weather is passed to the model. This function can be used to fetch weather data from your own data source. The function_call argument is set to force the model to use this function.

You can find more examples in the examples/src/models/chat/integration_openai.ts file in the LangChainJS repository.

Also, the LangChainJS framework does support integration with Azure OpenAI chat completion model. You can create a new instance of the OpenAI chat model and specify the model version and temperature as shown in the code snippet below:

import { ChatOpenAI } from "langchain/chat_models/openai";

const model = new ChatOpenAI({
  modelName: "gpt-4-0613",
  temperature: 0,
});

However, the context shared does not provide information about how to authenticate requests to the OpenAI API. You might need to refer to the Azure OpenAI documentation for that.

I hope this helps! If you have any other questions, feel free to ask.

Best,
Dosu

Sources

• examples/src/models/chat/integration_openai.ts
• langchain/src/chains/openai_functions/tests/openapi.int.test.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.

[nipun-threestops]

this answer is not related to azure openai api use your own data.Can I get more precise answer?

[Elindo586]

I just gave up asking their bot.

If you want just a bot to answer from a given questions I would look for these:

import { ChatOpenAI } from "langchain/chat_models/openai";
import { ConversationalRetrievalQAChain } from "langchain/chains";
import { TextLoader } from "langchain/document_loaders/fs/text";
import { CharacterTextSplitter } from "langchain/text_splitter";
import { OpenAIEmbeddings } from "langchain/embeddings/openai";

^^^ Then you gonna need to work with your vector store to add data...

If you have the time I would better go straight to OpenAI and see how they do it there...

I am currently working with Langchain for a bot with data and memory... and so far the bot just has me going in circles.

For now I am just going to answer question from data.... later on I am going to implement a better solution without Langchain..

[nipun-threestops]

@Elindo586 true,their bot is not helpful. Vector store as datasource is not exactly what I needed,In azure openai serivices there is a chat playground where we can add datasources to chat with.I have implemented REST api for this in my webapp, now for optimisation purpose trying to integrate with langhcain.so there is an option to connect with azure open ai deployment as LLM in langchain docs,but no option to configure the datasources.

import { OpenAI } from "langchain/llms/openai";

const model = new OpenAI({
  temperature: 0.9,
  azureOpenAIApiKey: "YOUR-API-KEY",
  azureOpenAIApiVersion: "YOUR-API-VERSION",
  azureOpenAIApiInstanceName: "{MY_INSTANCE_NAME}",
  azureOpenAIApiDeploymentName: "{DEPLOYMENT_NAME}",
});
const res = await model.call(
  "What would be a good company name a company that makes colorful socks?"
);
console.log({ res });

ref : https://js.langchain.com/docs/modules/model_io/models/llms/integrations/azure

[calls9-amirbraham]

@nipun-threestops any updates on this ?