Upload 3 files

app.py

@@ -0,0 +1,91 @@
+import streamlit as st
+from PyPDF2 import PdfReader
+
+from langchain.embeddings.openai import OpenAIEmbeddings
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.vectorstores import FAISS
+from langchain.chains.question_answering import load_qa_chain
+from langchain.llms import OpenAI
+import os 
+
+
+
+# Set API keys
+os.environ["OPENAI_API_KEY"] = "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"
+os.environ["SERPAPI_API_KEY"] = "28c2445d1bfe7530595be6fbc858b1968d776af69c6034aa5feda50deab4b990"
+
+
+# PDF processing
+pdfreader = PdfReader('XYZ_contract_pdf_Sumit Yenugwar.pdf')
+raw_text = ''
+for i, page in enumerate(pdfreader.pages):
+    content = page.extract_text()
+    if content:
+        raw_text += content
+
+# Text splitting
+text_splitter = CharacterTextSplitter(
+    separator="\n",
+    chunk_size=4000,
+    chunk_overlap=1000,
+    length_function=len,
+)
+texts = text_splitter.split_text(raw_text)
+
+# Download embeddings from OpenAI
+embeddings = OpenAIEmbeddings()
+
+
+# Create document search
+document_search = FAISS.from_texts(texts, embeddings)
+
+#########################################################
+document_search.save_local("faiss_index")
+
+#below lines loads the vectorized data that was saved in previous code line
+new_document_search = FAISS.load_local("faiss_index", embeddings)
+ 
+
+##################################
+
+# Load QA chain
+chain = load_qa_chain(OpenAI(), chain_type="stuff")
+
+with st.sidebar:
+    st.title('🤗💬 LLM Chat APP')
+    st.markdown('''
+    ## About
+    This app is an LLM-powered chatbot built using:
+    - [Streamlit](https://streamlit.io/)
+    - [LangChain](https://python.langchain.com/)
+    - [OpenAI](https://platform.openai.com/docs/models) LLM model
+ 
+    ''')
+    st.markdown("<br>", unsafe_allow_html=True)  # Add vertical space
+    st.write('Made with ❤️ by [Prompt Engineer](https://www.youtube.com/watch?v=M4mc-z_K1NU&list=PLUTApKyNO6MwrOioHGaFCeXtZpchoGv6W)')
+
+# Streamlit app
+def main():
+    st.title("DAMA-Data Management body of knowledge")
+
+    # Text input area
+    user_input = st.text_area("Enter your MCQ question ",height=150)
+
+    # Button to trigger model inference
+    if st.button("Get Answer"):
+        # Combine user input with the prompt and query
+        prompt_query = f"you have provided with MCQ question and its option as a chatbot model: {user_input}"
+        text_query = prompt_query + user_input
+
+        # Perform similarity search
+        docs = new_document_search.similarity_search(text_query)
+
+        # Run the model with the combined text and query
+        model_answer = chain.run(input_documents=docs, question=user_input)
+
+        # Display the model's answer
+        st.text_area("Model Answer:", value=model_answer)
+
+# Run the Streamlit app
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,6 @@
+streamlit==1.29.0
+PyPDF2==3.0.1
+langchain==0.0.345  # Replace with the correct version if available
+openai==1.3.7  # Replace with the correct version if available
+numpy==1.26.2
+tiktoken==0.5.2