Delete lab/lacks_persistence.py

lab/lacks_persistence.py

@@ -1,151 +0,0 @@
-import os
-import requests
-import streamlit as st
-from langchain.chains import SequentialChain, LLMChain
-from langchain.prompts import PromptTemplate
-from langchain_groq import ChatGroq
-from langchain.document_loaders import PDFPlumberLoader
-from langchain_experimental.text_splitter import SemanticChunker
-from langchain_huggingface import HuggingFaceEmbeddings
-from langchain_chroma import Chroma
-from prompts import rag_prompt, relevancy_prompt, relevant_context_picker_prompt, response_synth
-
-
-# Set API Keys
-os.environ["GROQ_API_KEY"] = st.secrets.get("GROQ_API_KEY", "")
-
-# Load LLM models
-llm_judge = ChatGroq(model="deepseek-r1-distill-llama-70b")
-rag_llm = ChatGroq(model="mixtral-8x7b-32768")
-
-llm_judge.verbose = True
-rag_llm.verbose = True
-
-st.title("❓")
-
-# Step 1: Choose PDF Source
-#### Initialize pdf_path
-pdf_path = None
-pdf_source = st.radio("Upload or provide a link to a PDF:", ["Upload a PDF file", "Enter a PDF URL"], index=0, horizontal=True)
-
-
-if pdf_source == "Upload a PDF file":
-    uploaded_file = st.file_uploader("Upload your PDF file", type="pdf")
-    if uploaded_file:
-        with open("temp.pdf", "wb") as f:
-            f.write(uploaded_file.getbuffer())
-        pdf_path = "temp.pdf"
-
-elif pdf_source == "Enter a PDF URL":
-    pdf_url = st.text_input("Enter PDF URL:")
-    if pdf_url:
-        with st.spinner("Downloading PDF..."):
-            try:
-                response = requests.get(pdf_url)
-                if response.status_code == 200:
-                    with open("temp.pdf", "wb") as f:
-                        f.write(response.content)
-                    pdf_path = "temp.pdf"
-                    st.success("✅ PDF Downloaded Successfully!")
-                else:
-                    st.error("❌ Failed to download PDF. Check the URL.")
-                    pdf_path = None
-            except Exception as e:
-                st.error(f"Error downloading PDF: {e}")
-                pdf_path = None
-else:
-    pdf_path = None
-
-# Step 2: Process PDF
-if pdf_path:
-    with st.spinner("Loading PDF..."):
-        loader = PDFPlumberLoader(pdf_path)
-        docs = loader.load()
-    
-    st.success(f"✅ **PDF Loaded!** Total Pages: {len(docs)}")
-
-    # Step 3: Chunking
-    with st.spinner("Chunking the document..."):
-        model_name = "nomic-ai/modernbert-embed-base"
-        embedding_model = HuggingFaceEmbeddings(model_name=model_name, model_kwargs={'device': 'cpu'})
-        text_splitter = SemanticChunker(embedding_model)
-        documents = text_splitter.split_documents(docs)
-
-    st.success(f"✅ **Document Chunked!** Total Chunks: {len(documents)}")
-
-    # Step 4: Setup Vectorstore
-    with st.spinner("Creating vector store..."):
-        vector_store = Chroma(
-            collection_name="deepseek_collection",
-            collection_metadata={"hnsw:space": "cosine"},
-            embedding_function=embedding_model
-        )
-        vector_store.add_documents(documents)
-        num_documents = len(vector_store.get()["documents"])
-
-    st.success(f"✅ **Vector Store Created!** Total documents stored: {num_documents}")
-
-    # Step 5: Query Input
-    query = st.text_input("🔍 Enter a Query:")
-    if query:
-        with st.spinner("Retrieving relevant contexts..."):
-            retriever = vector_store.as_retriever(search_type="similarity", search_kwargs={"k": 5})
-            contexts = retriever.invoke(query)
-            context_texts = [doc.page_content for doc in contexts]
-
-        st.success(f"✅ **Retrieved {len(context_texts)} Contexts!**")
-        for i, text in enumerate(context_texts, 1):
-            st.write(f"**Context {i}:** {text[:500]}...")
-
-        # Step 6: Context Relevancy Checker
-        with st.spinner("Evaluating context relevancy..."):
-            context_relevancy_checker_prompt = PromptTemplate(
-                input_variables=["retriever_query", "context"], template=relevancy_prompt
-            )
-            context_relevancy_chain = LLMChain(llm=llm_judge, prompt=context_relevancy_checker_prompt, output_key="relevancy_response")
-            relevancy_response = context_relevancy_chain.invoke({"context": context_texts, "retriever_query": query})
-
-        st.subheader("🟥 Context Relevancy Evaluation")
-        st.json(relevancy_response['relevancy_response'])
-
-        # Step 7: Selecting Relevant Contexts
-        with st.spinner("Selecting the most relevant contexts..."):
-            relevant_prompt = PromptTemplate(
-                input_variables=["relevancy_response"], template=relevant_context_picker_prompt
-            )
-            pick_relevant_context_chain = LLMChain(llm=llm_judge, prompt=relevant_prompt, output_key="context_number")
-            relevant_response = pick_relevant_context_chain.invoke({"relevancy_response": relevancy_response['relevancy_response']})
-
-        st.subheader("🟦 Pick Relevant Context Chain")
-        st.json(relevant_response['context_number'])
-
-        # Step 8: Retrieving Context for Response Generation
-        with st.spinner("Retrieving final context..."):
-            context_prompt = PromptTemplate(
-                input_variables=["context_number", "context"], template=response_synth
-            )
-            relevant_contexts_chain = LLMChain(llm=llm_judge, prompt=context_prompt, output_key="relevant_contexts")
-            final_contexts = relevant_contexts_chain.invoke({"context_number": relevant_response['context_number'], "context": context_texts})
-
-        st.subheader("🟥 Relevant Contexts Extracted")
-        st.json(final_contexts['relevant_contexts'])
-
-        # Step 9: Generate Final Response
-        with st.spinner("Generating the final answer..."):
-            final_prompt = PromptTemplate(
-                input_variables=["query", "context"], template=rag_prompt
-            )
-            response_chain = LLMChain(llm=rag_llm, prompt=final_prompt, output_key="final_response")
-            final_response = response_chain.invoke({"query": query, "context": final_contexts['relevant_contexts']})
-
-        st.subheader("🟥 RAG Final Response")
-        st.success(final_response['final_response'])
-
-        # Step 10: Display Workflow Breakdown
-        st.subheader("🔍 **Workflow Breakdown:**")
-        st.json({
-            "Context Relevancy Evaluation": relevancy_response["relevancy_response"],
-            "Relevant Contexts": relevant_response["context_number"],
-            "Extracted Contexts": final_contexts["relevant_contexts"],
-            "Final Answer": final_response["final_response"]
-        })