watsonx question answering

qa.py

# Import necessary libraries
import os
from ibm_watson_machine_learning.foundation_models import Model
from ibm_watson_machine_learning.metanames import GenTextParamsMetaNames as GenParams
from ibm_watson_machine_learning.foundation_models.utils.enums import ModelTypes, DecodingMethods

def get_model(model_type, api_key, project_id, max_tokens, min_tokens, decoding, temperature, top_k, top_p):
    generate_params = {
        GenParams.MAX_NEW_TOKENS: max_tokens,
        GenParams.MIN_NEW_TOKENS: min_tokens,
        GenParams.DECODING_METHOD: decoding,
        GenParams.TEMPERATURE: temperature,
        GenParams.TOP_K: top_k,
        GenParams.TOP_P: top_p,
    }

    model = Model(
        model_id=model_type,
        params=generate_params,
        credentials={
            "apikey": api_key,
            "url": "https://us-south.ml.cloud.ibm.com"
        },
        project_id=project_id
    )

    return model

def qa(question):
    # Fetch environment variables
    api_key = os.getenv("WATSONX_API_KEY")
    project_id = os.getenv("PROJECT_ID")

    if not api_key or not project_id:
        raise ValueError("API key and Project ID must be set in environment variables")

    # Specify model parameters
    model_type = "meta-llama/llama-3-8b-instruct"
    max_tokens = 100
    min_tokens = 50
    top_k = 50
    top_p = 1
    decoding = DecodingMethods.GREEDY
    temperature = 0.7

    # Get the WatsonX model
    model = get_model(model_type, api_key, project_id, max_tokens, min_tokens, decoding, temperature, top_k, top_p)
    complete_prompt = create_prompt(question)

    # Print the prompt for review
    print("----------------------------------------------------------------------------------------------------")
    print(f"*** Prompt: {complete_prompt} ***")
    print("----------------------------------------------------------------------------------------------------")

    generated_response = model.generate(prompt=complete_prompt)
    response_text = generated_response['results'][0]['generated_text']

    # Remove trailing white spaces
    return response_text.strip()

def create_prompt(question):
  prompt = f"""<|begin_of_text|>
<|start_header_id|>system<|end_header_id|>
You are a large language model trained to be informative and comprehensive. 
I am a chatbot assistant trying to answer questions for users in a proper and informative way. 
The user asked: "{question}"
Using your knowledge and the provided URL (if available), answer the user's question in a comprehensive and informative way, avoiding any informal language or slang.
<|end_of_text|>"""
  return prompt
qa("How are you")