In the vast world of artificial intelligence, language processing is a captivating and complex field. It requires robust frameworks and innovative solutions to comprehend and generate human language. Excitingly, MinsdDB's integration with LangChain combines powerful in-database AI capabilities and a cutting-edge language processing framework for developing applications powered by large language models (LLMs). Together, they empower developers, data scientists, and businesses to effortlessly interact with their databases, unlocking new possibilities for data exploration, analysis, and decision-making.

‍

LangChain models address the limitations of working directly with large language models (such as the GPT family from OpenAI) by providing better control and accuracy in text generation. LangChain allows users to connect multiple large language models in a logical way using chains, leveraging the strengths of each model and expanding the capabilities of language processing systems. It also enables the usage of mechanisms like memories and output parsers, which can improve the robustness of LLM-based applications. By employing chains, memories, and output parsers, LangChain empowers users to create, among other things, so-called “agents” that emerge out of forcing LLMs into a structured loop to achieve reasoning. These agents can understand and interact with data, making communication with data more efficient and intuitive.

‍

In this blog post, we delve into the transformative potential of MinsdDB's integration with LangChain starting from natural language understanding by the LangChain model to communicating with your database.

Communicate with your Database using LangChain and MindsDB

The ability to seamlessly communicate with databases is a fundamental requirement for numerous applications and industries. With the powerful combination of LangChain and MindsDB, this essential task becomes not only efficient but also remarkably intuitive.

‍

Here is how to create a LangChain model in MindsDB to accomplish the aforementioned tasks:

CREATE MODEL tool_based_agent
PREDICT completion
USING
    engine = 'langchain',
    prompt_template = 'Answer the users input in a helpful way: {{input}}';

This model is going to act as an agent between you and your data. Every time we interact with the model, we pass the user’s input in the {{input}} variable.

Describing the Data

In the ever-evolving landscape of data-driven decision-making, the ability to accurately describe and understand complex datasets is of paramount importance.

‍

Here is how to inquire about data descriptions from the agent:

SELECT input, completion
FROM tool_based_agent
WHERE input = 'Could you describe the `mysql_demo_db.house_sales` table please?'
USING
    verbose = True,
    tools = [],
    max_iterations = 10;

We ask the agent to describe the house_sales table available in the connected data source named mysql_demo_db.

‍

And here is the answer:

The `mysql_demo_db.house_sales` table has four columns: 
    `saledate` (text), 
    `house_price_moving_average` (integer), 
    `type` (text), 
    and `bedrooms` (integer).

The agent uses its Metadata tool to fetch information about the table. Then it writes an answer back to the user.

Analyzing the Data

In the realm of data analysis, the ability to extract meaningful insights from vast amounts of information is a game-changer for businesses and researchers alike. We present here the transformative potential of utilizing LangChain and MindsDB for data analysis.

‍

Let's prompt our agent to perform data analysis:

SELECT input, completion
FROM tool_based_agent
WHERE input = 'I want to know the average number of beds in the downtown neighbourhood as per the `mysql_demo_db.home_rentals` table'
USING
    verbose = True,
    tools = [],
    max_iterations = 10;

Here the agent must figure out a way to find out the average number of beds in the downtown neighbourhood using the home_rentals table available in the connected data source named mysql_demo_db.

‍

Using the Metadata tool, the agent comes up with this query:

SELECT AVG(number_of_rooms)
FROM mysql_demo_db.home_rentals
WHERE neighborhood = 'downtown';

It returns 1.6, which is written back to the user as below:

The average number of beds in the downtown neighbourhood 
as per the `mysql_demo_db.home_rentals` table is 1.6.

To see more examples of how to describe, analyze, retrieve, and insert data using LangChain and MindsDB, visit our documentation here.

What’s Next

If you are already familiar with MinsdDB, we encourage you to try its new integration with the LangChain ML engine, as it enables you to retrieve insights and manipulate your data through natural language queries.

‍

And if you are new to MindsDB, go ahead and create a demo account to explore MindsDB Cloud. Join our Slack community to ask questions and share feedback.

‍