refactor the generator code

app.py

@@ -1,176 +1,14 @@
 import streamlit as st
 from utils.retriever import retrieve_paragraphs
+from utils.generator import build_messages, _call_llm
 import ast
 import time
 import asyncio
 import logging
 import logging
 logging.basicConfig(level=logging.INFO)
-import os
-import configparser
 
 
-def getconfig(configfile_path: str):
-    """
-    Read the config file
-    Params
-    ----------------
-    configfile_path: file path of .cfg file
-    """
-    config = configparser.ConfigParser()
-    try:
-        config.read_file(open(configfile_path))
-        return config
-    except:
-        logging.warning("config file not found")
-
-# ---------------------------------------------------------------------
-# Provider-agnostic authentication and configuration
-# ---------------------------------------------------------------------
-
-def get_auth(provider: str) -> dict:
-    """Get authentication configuration for different providers"""
-    auth_configs = {
-        "openai": {"api_key": os.getenv("OPENAI_API_KEY")},
-        "huggingface": {"api_key": os.getenv("HF_TOKEN")},
-        "anthropic": {"api_key": os.getenv("ANTHROPIC_API_KEY")},
-        "cohere": {"api_key": os.getenv("COHERE_API_KEY")},
-    }
-    
-    if provider not in auth_configs:
-        raise ValueError(f"Unsupported provider: {provider}")
-    
-    auth_config = auth_configs[provider]
-    api_key = auth_config.get("api_key")
-    
-    if not api_key:
-        raise RuntimeError(f"Missing API key for provider '{provider}'. Please set the appropriate environment variable.")
-    
-    return auth_config
-
-# ---------------------------------------------------------------------
-# Model / client initialization (non exaustive list of providers)
-# ---------------------------------------------------------------------
-
-config = getconfig("model_params.cfg")
-
-PROVIDER = config.get("generator", "PROVIDER")
-MODEL = config.get("generator", "MODEL")
-MAX_TOKENS = int(config.get("generator", "MAX_TOKENS"))
-TEMPERATURE = float(config.get("generator", "TEMPERATURE"))
-INFERENCE_PROVIDER = config.get("generator", "INFERENCE_PROVIDER")
-ORGANIZATION = config.get("generator", "ORGANIZATION")
-
-# Set up authentication for the selected provider
-auth_config = get_auth(PROVIDER)
-
-
-from langchain_core.messages import SystemMessage, HumanMessage
-from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
-
-def build_messages(question: str, context: str) -> list:
-    """
-    Build messages in LangChain format.
-    
-    Args:
-        question: The user's question
-        context: The relevant context for answering
-        
-    Returns:
-        List of LangChain message objects
-    """
-    system_content = (
-        """
-        You are an expert assistant. Your task is to generate accurate, helpful responses using only the 
-        information contained in the "CONTEXT" provided.
-        Instructions:
-        - Answer based only on provided context: Use only the information present in the retrieved_paragraphs below. Do not use any external knowledge or make assumptions beyond what is explicitly stated.
-        - Language matching: Respond in the same language as the user's query.
-        - Handle missing information: If the retrieved paragraphs do not contain sufficient information to answer the query, respond with "I don't know" or equivalent in the query language. If information is incomplete, state what you know and acknowledge limitations.
-        - Be accurate and specific: When information is available, provide clear, specific answers. Include relevant details, useful facts, and numbers from the context.
-        - Stay focused: Answer only what is asked. Do not provide additional information not requested.
-        - Structure your response effectively:
-                * Do not just summarize each passage one by one. Group your summaries to highlight the key parts in the explanation.
-                * Use bullet points and lists when it makes sense to improve readability.
-                * You do not need to use every passage. Only use the ones that help answer the question.
-        - Format your response properly: Use markdown formatting (bullet points, numbered lists, headers) to make your response clear and easy to read. Example: <br> for linebreaks 
-        
-        Input Format:
-        - Query: {query}
-        - Retrieved Paragraphs: {retrieved_paragraphs}
-        Generate your response based on these guidelines.
-        """
-    )
-    
-    user_content = f"### CONTEXT\n{context}\n\n### USER QUESTION\n{question}"
-    
-    return [
-        SystemMessage(content=system_content),
-        HumanMessage(content=user_content)
-    ]
-def get_chat_model():
-    """Initialize the appropriate LangChain chat model based on provider"""
-    common_params = {
-        "temperature": TEMPERATURE,
-        "max_tokens": MAX_TOKENS,
-    }
-    
-    # if PROVIDER == "openai":
-    #     return ChatOpenAI(
-    #         model=MODEL,
-    #         openai_api_key=auth_config["api_key"],
-    #         **common_params
-    #     )
-    # elif PROVIDER == "anthropic":
-    #     return ChatAnthropic(
-    #         model=MODEL,
-    #         anthropic_api_key=auth_config["api_key"],
-    #         **common_params
-    #     )
-    # elif PROVIDER == "cohere":
-    #     return ChatCohere(
-    #         model=MODEL,
-    #         cohere_api_key=auth_config["api_key"],
-    #         **common_params
-    #     )
-    if PROVIDER == "huggingface":
-        # Initialize HuggingFaceEndpoint with explicit parameters
-        llm = HuggingFaceEndpoint(
-            repo_id=MODEL,
-            huggingfacehub_api_token=auth_config["api_key"],
-            task="text-generation",
-            provider=INFERENCE_PROVIDER,     
-            server_kwargs={"bill_to": ORGANIZATION},
-            temperature=TEMPERATURE,
-            max_new_tokens=MAX_TOKENS
-        )
-        return ChatHuggingFace(llm=llm)
-    else:
-        raise ValueError(f"Unsupported provider: {PROVIDER}")
-
-# Initialize provider-agnostic chat model
-chat_model = get_chat_model()
-
-async def _call_llm(messages: list) -> str:
-    """
-    Provider-agnostic LLM call using LangChain.
-    
-    Args:
-        messages: List of LangChain message objects
-        
-    Returns:
-        Generated response content as string
-    """
-    try:
-        # Use async invoke for better performance
-        response = await chat_model.ainvoke(messages)
-        logging.info(f"answer: {response.content}")
-        return response.content
-        #return response.content.strip()
-    except Exception as e:
-        logging.exception(f"LLM generation failed with provider '{PROVIDER}' and model '{MODEL}': {e}")
-        raise
-
 
 def chat_response(query):
     """Generate chat response based on method and inputs"""
@@ -186,8 +24,6 @@ def chat_response(query):
     
         messages = build_messages(query, context_retrieved_lst)
         answer = asyncio.run(_call_llm(messages))
-
-
         return answer
 
                 
@@ -214,3 +50,4 @@ if not query.strip():
     st.stop()
 else:
     st.write(chat_response(query))
+    

utils/generator.py

@@ -0,0 +1,128 @@
+from utils.utils import getconfig, get_auth
+from langchain_core.messages import SystemMessage, HumanMessage
+from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
+import logging
+# ---------------------------------------------------------------------
+# Model / client initialization (non exaustive list of providers)
+
+config = getconfig("model_params.cfg")
+# Reading Params
+PROVIDER = config.get("generator", "PROVIDER")
+MODEL = config.get("generator", "MODEL")
+MAX_TOKENS = int(config.get("generator", "MAX_TOKENS"))
+TEMPERATURE = float(config.get("generator", "TEMPERATURE"))
+INFERENCE_PROVIDER = config.get("generator", "INFERENCE_PROVIDER")
+ORGANIZATION = config.get("generator", "ORGANIZATION")
+
+# Set up authentication for the selected provider
+auth_config = get_auth(PROVIDER)
+
+
+def get_chat_model():
+    """Initialize the appropriate LangChain chat model based on provider"""
+    common_params = {
+        "temperature": TEMPERATURE,
+        "max_tokens": MAX_TOKENS,
+    }
+    
+    # if PROVIDER == "openai":
+    #     return ChatOpenAI(
+    #         model=MODEL,
+    #         openai_api_key=auth_config["api_key"],
+    #         **common_params
+    #     )
+    # elif PROVIDER == "anthropic":
+    #     return ChatAnthropic(
+    #         model=MODEL,
+    #         anthropic_api_key=auth_config["api_key"],
+    #         **common_params
+    #     )
+    # elif PROVIDER == "cohere":
+    #     return ChatCohere(
+    #         model=MODEL,
+    #         cohere_api_key=auth_config["api_key"],
+    #         **common_params
+    #     )
+    if PROVIDER == "huggingface":
+        # Initialize HuggingFaceEndpoint with explicit parameters
+        llm = HuggingFaceEndpoint(
+            repo_id=MODEL,
+            huggingfacehub_api_token=auth_config["api_key"],
+            task="text-generation",
+            provider=INFERENCE_PROVIDER,     
+            server_kwargs={"bill_to": ORGANIZATION},
+            temperature=TEMPERATURE,
+            max_new_tokens=MAX_TOKENS
+        )
+        return ChatHuggingFace(llm=llm)
+    else:
+        raise ValueError(f"Unsupported provider: {PROVIDER}")
+
+# Initialize provider-agnostic chat model
+chat_model = get_chat_model()
+
+#------------------------------------Define Prompt -----------------------------------------
+
+
+def build_messages(question: str, context: str) -> list:
+    """
+    Build messages in LangChain format.
+    
+    Args:
+        question: The user's question
+        context: The relevant context for answering
+        
+    Returns:
+        List of LangChain message objects
+    """
+    system_content = (
+        """
+        You are an expert assistant. Your task is to generate accurate, helpful responses using only the 
+        information contained in the "CONTEXT" provided.
+        Instructions:
+        - Answer based only on provided context: Use only the information present in the retrieved_paragraphs below. Do not use any external knowledge or make assumptions beyond what is explicitly stated.
+        - Language matching: Respond in the same language as the user's query.
+        - Handle missing information: If the retrieved paragraphs do not contain sufficient information to answer the query, respond with "I don't know" or equivalent in the query language. If information is incomplete, state what you know and acknowledge limitations.
+        - Be accurate and specific: When information is available, provide clear, specific answers. Include relevant details, useful facts, and numbers from the context.
+        - Stay focused: Answer only what is asked. Do not provide additional information not requested.
+        - Structure your response effectively:
+                * Do not just summarize each passage one by one. Group your summaries to highlight the key parts in the explanation.
+                * Use bullet points and lists when it makes sense to improve readability.
+                * You do not need to use every passage. Only use the ones that help answer the question.
+        - Format your response properly: Use markdown formatting (bullet points, numbered lists, headers) to make your response clear and easy to read. Example: <br> for linebreaks 
+        
+        Input Format:
+        - Query: {query}
+        - Retrieved Paragraphs: {retrieved_paragraphs}
+        Generate your response based on these guidelines.
+        """
+    )
+    
+    user_content = f"### CONTEXT\n{context}\n\n### USER QUESTION\n{question}"
+    
+    return [
+        SystemMessage(content=system_content),
+        HumanMessage(content=user_content)
+    ]
+
+#--------------------------------Get the async response ---------------------------------------------
+async def _call_llm(messages: list) -> str:
+    """
+    Provider-agnostic LLM call using LangChain.
+    
+    Args:
+        messages: List of LangChain message objects
+        
+    Returns:
+        Generated response content as string
+    """
+    try:
+        # Use async invoke for better performance
+        response = await chat_model.ainvoke(messages)
+        logging.info(f"answer: {response.content}")
+        return response.content
+        #return response.content.strip()
+    except Exception as e:
+        logging.exception(f"LLM generation failed with provider '{PROVIDER}' and model '{MODEL}': {e}")
+        raise
+

utils/utils.py

@@ -0,0 +1,41 @@
+import os
+import configparser
+
+
+def getconfig(configfile_path: str):
+    """
+    Read the config file
+    Params
+    ----------------
+    configfile_path: file path of .cfg file
+    """
+    config = configparser.ConfigParser()
+    try:
+        config.read_file(open(configfile_path))
+        return config
+    except:
+        logging.warning("config file not found")
+
+# ---------------------------------------------------------------------
+# Provider-agnostic authentication and configuration
+# ---------------------------------------------------------------------
+
+def get_auth(provider: str) -> dict:
+    """Get authentication configuration for different providers"""
+    auth_configs = {
+        "openai": {"api_key": os.getenv("OPENAI_API_KEY")},
+        "huggingface": {"api_key": os.getenv("HF_TOKEN")},
+        "anthropic": {"api_key": os.getenv("ANTHROPIC_API_KEY")},
+        "cohere": {"api_key": os.getenv("COHERE_API_KEY")},
+    }
+    
+    if provider not in auth_configs:
+        raise ValueError(f"Unsupported provider: {provider}")
+    
+    auth_config = auth_configs[provider]
+    api_key = auth_config.get("api_key")
+    
+    if not api_key:
+        raise RuntimeError(f"Missing API key for provider '{provider}'. Please set the appropriate environment variable.")
+    
+    return auth_config