Ollama working setup for extracting the data from PDFs, and giving the answers based on contect in them