Use MongoDBQueryEngine to query Markdown files#

This notebook demonstrates the use of the MongoDBQueryEngine for retrieval-augmented question answering over documents. It shows how to set up the engine with Docling parsed Markdown files, and execute natural language queries against the indexed data.

The MongoDBQueryEngine integrates cloud MongoDB Atlas vector storage with LlamaIndex for efficient document retrieval.

%pip install llama-index==0.12.16
%pip install llama-index-vector-stores-mongodb==0.6.0
%pip install llama-index-embeddings-huggingface==0.5.2

Set up Open AI key for query engine retrieval#

import os

import autogen

config_list = autogen.config_list_from_json(env_or_file="../OAI_CONFIG_LIST")

assert len(config_list) > 0
print("models to use: ", [config_list[i]["model"] for i in range(len(config_list))])

# Put the OpenAI API key into the environment
os.environ["OPENAI_API_KEY"] = config_list[0]["api_key"]

Set up Mongo DB instance#

To use this notebook, you will need to have a MongoDB Atlas environment. For this notebook, we use a docker instance. please refer MongoDB: Create a Local Atlas Deployment with Docker for more info.

Some info that you need to get include: - MongoDB Connection String (URI)

from autogen.agentchat.contrib.rag.mongodb_query_engine import MongoDBQueryEngine

query_engine = MongoDBQueryEngine(
    connection_string="mongodb://localhost:27017/?directConnection=true",
    database_name="vector_db",
    collection_name="test_collection",
)

Here we can see the default collection name in the vector store, this is where all documents will be ingested. When creating the MongoDBQueryEngine you can specify a collection_name and database_name to ingest into.

print(query_engine.get_collection_name())

Initialize DB and ingest documents#

Let’s ingest a document and query it.
init_db will overwrite the existing collection with the same name.

input_dir = "../test/agents/experimental/document_agent/pdf_parsed/"  # Update to match your input directory

input_docs = [input_dir + "nvidia_10k_2024.md"]  # Update to match your input documents
query_engine.init_db(new_doc_paths_or_urls=input_docs)

If the given collection already has the document you need, you can use connect_db to avoid overwriting the existing collection.

query_engine.connect_db()

question = "What is Nvidia's revenue in 2024?"
answer = query_engine.query(question)
print(answer)

Great, we got the data we needed. Now, let’s add another document.

new_docs = [input_dir + "Toast_financial_report.md"]
query_engine.add_docs(new_doc_paths_or_urls=new_docs)

And query again from the same database but this time for another corporate entity.

question = "What is the trading symbol for Toast"
answer = query_engine.query(question)
print(answer)