Update app.py

app.py

@@ -1,3 +1,75 @@
+import subprocess
+import sys
+import os
+import uuid
+import json
+from pathlib import Path
+import gradio as gr
+from dotenv import load_dotenv
+from langchain_community.embeddings.sentence_transformer import SentenceTransformerEmbeddings
+from langchain_community.vectorstores import Chroma
+from huggingface_hub import login
+import openai
+
+def install_packages():
+    packages = ["openai==0.28", "langchain_community", "sentence-transformers", "chromadb", "huggingface_hub", "python-dotenv", "numpy", "scipy", "scikit-learn"]
+    for package in packages:
+        subprocess.check_call([sys.executable, "-m", "pip", "install", "--upgrade", package])
+
+install_packages()
+
+# Load environment variables from .env file
+load_dotenv()
+
+# Get API tokens from environment variables
+openai.api_key = os.getenv("OPENAI_API_KEY")  # Ensure OPENAI_API_KEY is in your .env file
+hf_token = os.getenv("hf_token")
+
+if not hf_token:
+    raise ValueError("Hugging Face token is missing. Please set 'hf_token' as an environment variable.")
+
+# Log in to Hugging Face
+login(hf_token)
+print("Logged in to Hugging Face successfully.")
+
+# Set up embeddings and vector store
+embeddings = SentenceTransformerEmbeddings(model_name="thenlper/gte-large")
+collection_name = 'report-10k-2024'
+
+vectorstore_persisted = Chroma(
+    collection_name=collection_name,
+    persist_directory='./report_10kdb',
+    embedding_function=embeddings
+)
+
+# Set up the retriever
+retriever = vectorstore_persisted.as_retriever(
+    search_type='similarity',
+    search_kwargs={'k': 5}
+)
+
+# Define Q&A system messages
+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.
+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.
+Here is an example of how to structure your response:
+Answer:
+[Answer]
+Source:
+[Source]
+"""
+
+qna_user_message_template = """
+###Context
+Here are some documents that are relevant to the question.
+{context}
 """
 
 # Define the predict function
@@ -89,7 +161,3 @@ with gr.Blocks(theme="gr.themes.Monochrome()") as demo:
 
 demo.queue()
 demo.launch()
-
-
-
-