chat-with-loandecisionservice

Introduction

The integration of the Langchain project with ODM, a cutting-edge decision management product, signifies a significant leap forward in the realms of language processing and data-driven decision-making. By combining Langchain's advanced language analysis technology with the powerful capabilities of ODM, users can unlock a multitude of benefits that revolutionize their decision-making processes. This integration offers a seamless and robust solution for businesses looking to harness the power of multilingual data, optimize operations, and drive strategic outcomes. In the following lines, we will explore five compelling advantages that arise from this integration.

Pre-requisites

• Python 3.8 or higher
• docker
• An OpenAI account

Setup Pre-Requisites

Create a virtual env and install the Python package

python3 -m venv ~/llmdc
~/llmdc/bin/activate
pip install -r requirements.txt

Run the demo

Run the ODM for Developers docker image

Open a terminal

docker run -e USERS_PASSWORD=odmAdmin -e LICENSE=accept -p 9060:9060 -p 9443:9443  -m 2048M --memory-reservation 2048M  -e SAMPLE=true icr.io/cpopen/odm-k8s/odm:8.11

Run the Chat application

Open a new terminal

export ODM_ADMIN_PASSWORD=odmAdmin
export OPENAI_API_KEY=<OPENAI_API_KEY>
python3 app-loan.py

Then open a browser to this url : http://127.0.0.1:7860

Then you can ask queston to the chat bot such as:

• My name is Joe. I want to apply for a loan of $100,000 with a credit score of 210 and a duration of 100. Is it possible?
• Will it be possible to secure a loan of $100,000 with a credit score of 210 for a period of 10 years?

Code explain

1. Extract from the Decision Server the Ruleset definition.

def decisionServiceOperation():

    spec = OpenAPISpec.from_url("http://localhost:9060/DecisionService/rest/v1/mydeployment/1.0/Miniloan_ServiceRuleset/1.0/OPENAPI?format=YAML")

    operation = APIOperation.from_openapi_spec(spec, '/mydeployment/1.0/Miniloan_ServiceRuleset/1.0', "post")
    return operation
operation=decisionServiceOperation()

2. Initialize the LLM Agent. Agent make a glue between the LLM and the backend throw LangChain

def initializeLLMAgent(operation):
    llm = OpenAI() # Load a Language Model
    # Manage ODM Basic Authentication 
    message = "odmAdmin:"+os.environ["ODM_ADMIN_PASSWORD"]
    message_bytes = message.encode('ascii')
    base64_bytes = base64.b64encode(message_bytes)
    base64_message = base64_bytes.decode('ascii')
    headers = {
        "Authorization": f"Basic "+base64_message
    }
    odm_requests_wrapper=RequestsWrapper(headers=headers)
    # Load the Decision Operation Swagger definitation in the LLM Model.
    chain = OpenAPIEndpointChain.from_api_operation(
                                                    operation, 
                                                    llm, 
                                                    verify=False,
                                                    requests=odm_requests_wrapper, 
                                                    verbose=True,
                                                    return_intermediate_steps=True)
    return chain
odm_agent = initializeLLMAgent(operation)

3. Interact with the Agent

 response['output'] = odm_agent(query) # Query is the user input.