Create app.py

app.py

@@ -0,0 +1,74 @@
+import streamlit as st
+import os
+from langchain_community.document_loaders import PyMuPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_openai import OpenAIEmbeddings, ChatOpenAI
+from langchain_community.vectorstores import Qdrant
+from langchain.prompts import ChatPromptTemplate
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.runnables import RunnablePassthrough
+from qdrant_client import QdrantClient
+
+# Set up API keys
+os.environ["OPENAI_API_KEY"] = st.secrets["OPENAI_API_KEY"]
+
+# PDF links
+pdf_links = [
+    "https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf",
+    "https://www.whitehouse.gov/wp-content/uploads/2022/10/Blueprint-for-an-AI-Bill-of-Rights.pdf"
+]
+
+# Load and process PDFs
+@st.cache_resource
+def load_and_process_pdfs(pdf_links):
+    documents = []
+    for link in pdf_links:
+        loader = PyMuPDFLoader(file_path=link)
+        documents.extend(loader.load())
+
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=500,
+        chunk_overlap=40,
+        length_function=len,
+        is_separator_regex=False
+    )
+
+    return text_splitter.split_documents(documents)
+
+# Set up Qdrant and embeddings
+@st.cache_resource
+def setup_vectorstore():
+    qdrant_client = QdrantClient(":memory:")
+    embeddings = OpenAIEmbeddings()
+    
+    COLLECTION_NAME = "AI_Ethics_Framework"
+    vector_store = Qdrant(
+        client=qdrant_client, 
+        collection_name=COLLECTION_NAME,
+        embeddings=embeddings
+    )
+    
+    # Check if collection exists, if not, create and populate it
+    collections = qdrant_client.get_collections().collections
+    if not any(collection.name == COLLECTION_NAME for collection in collections):
+        documents = load_and_process_pdfs(pdf_links)
+        vector_store.add_documents(documents)
+    
+    return vector_store
+
+# Create RAG pipeline
+@st.cache_resource
+def create_rag_pipeline(vector_store):
+    retriever = vector_store.as_retriever()
+    
+    template = """
+    You are an expert AI assistant with deep knowledge of business, technology, and entrepreneurship. Your task is to provide accurate, insightful answers based solely on the given context. Follow these guidelines:
+
+    1. Analyze the question carefully to understand the core information being sought.
+    2. Thoroughly examine the provided context, identifying key relevant information.
+    3. Formulate a clear, concise answer that directly addresses the question.
+    4. Use specific details and examples from the context to support your answer.
+    5. If the context doesn't contain sufficient information to fully answer the question, state this clearly and say,'I don't know'.
+    6. Do not introduce any information not present in the context.
+    7. If asked for an opinion or recommendation, base it strictly on insights from the context.
+    8. Use