- Node.js v18x and npm (If not installed, download from Node.js official website)
-
Clone the Repository:
git clone [email protected]:jsakeag/langchain-healthbot-new.git cd langchain-healthbot-new
-
Install Dependencies:
npm install
If needed:
pip install -r /path/to/requirements.txt
-
Set Up OpenAI API Key:
-
Obtain your OpenAI API key.
-
Create a .env file in the root directory and add your OpenAI key.
REACT_APP_OPEN_AI_API_KEY=your_openai_api_key
-
Start the Backend and Frontend Servers:
cd backend python3 app.pycd frontend npm start
- Q/A with our online KG: The default way option uses a neo4j knowledge graph we created ourselves.
Notice that the
chatbot_responseimport is fromchatbot_KG.pyin theapp.pyfile. Here is a code snippet from our chatbot_KG.py file:
kg = Neo4jGraph(
url="neo4j+s://4221dbdc.databases.neo4j.io",
username = "neo4j",
password="6wmPE5kyuqGAxoPBzZeB12Mb56gzaNpP9goIC-8qdAc",
database = 'neo4j'
)
neo4j_vector_store = Neo4jVector.from_existing_graph(
embedding=OpenAIEmbeddings(),
url='neo4j+s://4221dbdc.databases.neo4j.io',
username='neo4j',
password='6wmPE5kyuqGAxoPBzZeB12Mb56gzaNpP9goIC-8qdAc',
index_name='embeddings',
node_label='Information',
text_node_properties=['text'],
embedding_node_property='embedding'
)-
ChatGPT Clone: The other way to run this project is the ChatGPT default 4o mini model. This can be selected by setting the
chatbot_responseimport tochatbot.pyin theapp.pyfile. Here is a code snippet from our chatbot.py file:# Initialize OpenAI Chat model llm = ChatOpenAI( model_name="gpt-4o-mini", temperature=0.7, max_tokens=500, api_key=openai_api_key # Explicitly pass the API key here ) # Initialize memory for conversation memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True) # Create embedding model embeddings = OpenAIEmbeddings(api_key=openai_api_key) # Also pass the API key here # Initialize Chroma DB (vector database) vector_db = Chroma(embedding_function=embeddings, collection_name="my_collection",persist_directory="./my_chroma_db")
- If you are experiencing CORS errors Make sure you allow your IP to access the endpoint (look at your AWS lambda scripts)