Update app.py

app.py

@@ -1,38 +1,30 @@
 import subprocess
 import sys
-# Install openai if it is not already installed
-subprocess.check_call([sys.executable, "-m", "pip", "install", "openai"])
-# Install langchain_community if it is not already installed
-subprocess.check_call([sys.executable, "-m", "pip", "install", "langchain_community"])
-# Install sentence-transformers if it is not already installed
-subprocess.check_call([sys.executable, "-m", "pip", "install", "sentence-transformers"])
-# Install sentence-transformers if it is not already installed
-subprocess.check_call([sys.executable, "-m", "pip", "install", "chromadb"])
-subprocess.check_call([sys.executable, "-m", "pip", "install", "huggingface_hub"])
-from huggingface_hub import login
-login("RAG")
-#huggingface-cli login 
-import openai
 import os
 import uuid
 import json
+from pathlib import Path
+from dotenv import load_dotenv
+from huggingface_hub import login, CommitScheduler
 import gradio as gr
-#from openai import OpenAI
 from langchain_community.embeddings.sentence_transformer import SentenceTransformerEmbeddings
 from langchain_community.vectorstores import Chroma
-#from huggingface_hub import login
-#login("RAG")
-from huggingface_hub import CommitScheduler
-from pathlib import Path
-from dotenv import load_dotenv
+import openai
+
+# Install required libraries if not already installed
+subprocess.check_call([sys.executable, "-m", "pip", "install", "openai", "langchain_community", "sentence-transformers", "chromadb", "huggingface_hub", "python-dotenv"])
 
 # Load environment variables from .env file
 load_dotenv()
 
-# Set OpenAI API key
-openai.api_key = os.getenv("OPENAI_API_KEY")  # Make sure OPENAI_API_KEY is in your .env file
+# Login to Hugging Face using token from environment variables
+hf_token = os.getenv("HF_TOKEN")
+if not hf_token:
+    raise ValueError("Hugging Face token not found in environment variables. Set HF_TOKEN in your .env file.")
+login(hf_token)
 
-# Initialize OpenAI client
+# Set OpenAI API key from environment variables
+openai.api_key = os.getenv("OPENAI_API_KEY")  # Ensure OPENAI_API_KEY is in your .env file
 client = openai
 
 # Set up embeddings and vectorstore
@@ -50,7 +42,7 @@ retriever = vectorstore_persisted.as_retriever(
     search_kwargs={'k': 5}
 )
 
-# Set up logging
+# Define logging configuration
 log_file = Path("logs/") / f"data_{uuid.uuid4()}.json"
 log_folder = log_file.parent
 
@@ -64,26 +56,19 @@ scheduler = CommitScheduler(
 
 # Define the Q&A system message
 qna_system_message = """
-You are an AI assistant to help Finsights Grey Inc., an innovative financial technology firm, develop a Retrieval-Augmented Generation (RAG) system to automate the extraction, summarization, and analysis of information from 10-K reports. Your knowledge base was last updated in August 2023.
-User input will have the context required by you to answer user questions. This context will begin with the token: ###Context.
-The context contains references to specific portions of a 10-K report relevant to the user query.
-User questions will begin with the token: ###Question.
-Your response should only be about the question asked and the context provided.
-Answer only using the context provided.
-Do not mention anything about the context in your final answer.
-If the answer is not found in the context, it is very important for you to respond with "I don't know."
-Always quote the source when you use the context. Cite the relevant source at the end of your response under the section - Source:
-Do not make up sources. Use the links provided in the sources section of the context and nothing else. You are prohibited from providing other links/sources.
+You are an AI assistant helping Finsights Grey Inc., a financial technology firm, develop a Retrieval-Augmented Generation (RAG) system to automate extraction, summarization, and analysis of 10-K reports.
+Your knowledge base was last updated in August 2023.
+User questions will start with the token: ###Question.
+Answer only based on the provided context.
+If the answer is not found in the context, respond with "I don't know."
 """
 
 qna_user_message_template = """
 ###Context
 Here are some documents that are relevant to the question.
 {context}
+###Question
 {question}
-
-css
-Copy code
 """
 
 # Define the predict function
@@ -96,7 +81,10 @@ def predict(user_input, company):
     context_for_query = ".".join(context_list)
 
     # Create messages
-    prompt = [        {'role': 'system', 'content': qna_system_message},        {'role': 'user', 'content': qna_user_message_template.format(            context=context_for_query,            question=user_input        )}    ]
+    prompt = [
+        {'role': 'system', 'content': qna_system_message},
+        {'role': 'user', 'content': qna_user_message_template.format(context=context_for_query, question=user_input)}
+    ]
 
     # Get response from the LLM
     try:
@@ -105,29 +93,24 @@ def predict(user_input, company):
             messages=prompt,
             temperature=0
         )
-
         prediction = response.choices[0].message.content
-
     except Exception as e:
         prediction = str(e)
 
     # Log inputs and outputs to a local log file
     with scheduler.lock:
         with log_file.open("a") as f:
-            f.write(json.dumps(
-                {
-                    'user_input': user_input,
-                    'retrieved_context': context_for_query,
-                    'model_response': prediction
-                }
-            ))
+            f.write(json.dumps({
+                'user_input': user_input,
+                'retrieved_context': context_for_query,
+                'model_response': prediction
+            }))
             f.write("\n")
 
     return prediction
 
-
+# Define the prediction interface function
 def get_predict(question, company):
-    # Implement your prediction logic here
     company_map = {
         "AWS": "aws",
         "IBM": "IBM",
@@ -135,14 +118,13 @@ def get_predict(question, company):
         "Meta": "meta",
         "Microsoft": "msft"
     }
-    
     selected_company = company_map.get(company)
     if not selected_company:
         return "Invalid company selected"
     
     return predict(question, selected_company)
 
-# Set-up the Gradio UI
+# Set up the Gradio UI
 with gr.Blocks(theme="gradio/seafoam@>=0.0.1,<0.1.0") as demo:
     with gr.Row():
         company = gr.Radio(["AWS", "IBM", "Google", "Meta", "Microsoft"], label="Select a company")
@@ -158,4 +140,6 @@ with gr.Blocks(theme="gradio/seafoam@>=0.0.1,<0.1.0") as demo:
     )
 
 demo.queue()
-demo.launch()
+demo.launch()
+
+