Main. Uploaded all the files required to run this app

README.md

@@ -1,12 +1,13 @@
 ---
 title: LinkedIn Profile QA
-emoji: 😻
-colorFrom: yellow
-colorTo: green
+emoji: 🚀
+colorFrom: indigo
+colorTo: pink
 sdk: streamlit
-sdk_version: 1.33.0
+sdk_version: 1.32.2
 app_file: app.py
 pinned: false
+license: unknown
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,105 @@
+import streamlit as st
+from doc_loading import read_pdf_text
+from llm_functions import get_conversational_chain
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_openai import OpenAIEmbeddings
+from langchain.vectorstores import FAISS
+import warnings
+warnings.filterwarnings('ignore')
+
+from dotenv import load_dotenv
+load_dotenv()
+from dataclasses import dataclass
+
+secrets = st.secrets["openai"]
+from openai import OpenAI
+client = OpenAI(api_key=secrets["OPENAI_API_KEY"])
+
+
+st.set_page_config(page_title="Resume QA")
+st.header("Feel free to ask any question")
+
+st.write("Welcome!!")
+# st.session_state
+
+
+if "option" not in st.session_state:
+    option = st.selectbox("How are you going to input your document?",
+             ("Upload LinkedIn PDF", "Upload xml"))
+# Conditionally show components based on user's choice
+file_name = ""
+main_text = ""
+
+if option == "Upload LinkedIn PDF":
+    uploaded_file = st.file_uploader("Choose a PDF file", type=["pdf"], key="uploading_pdf")
+    if uploaded_file is not None:
+        # Process the uploaded file
+        st.write("PDF uploaded successfully!")
+        main_text = read_pdf_text(uploaded_file)
+
+else:
+    st.write("This functionality in under construction.")
+
+
+if option == "Upload LinkedIn PDF":
+    if uploaded_file is not None and uploaded_file.type != "application/pdf":
+        st.error("Please choose a PDF file only.")
+
+
+if len(main_text) > 0:
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=500,
+        chunk_overlap=50,
+        length_function=len,
+        is_separator_regex='\n'
+    )
+    texts = text_splitter.create_documents([main_text])
+
+    embeddings = OpenAIEmbeddings(api_key=secrets['OPENAI_API_KEY'])
+    # store in vector db
+    db = FAISS.from_documents(texts, embeddings)
+
+    hash_name = f"{option.replace(' ', '-')}"
+    db.save_local(f'faiss_{hash_name}_index')
+
+    if "messages" not in st.session_state:
+        st.session_state.messages = []
+
+    st.write("What's your question? (try being a bit detailed about the question)")
+
+    @dataclass
+    class Message:
+        actor: str
+        payload: str
+
+    USER = "user"
+    ASSISTANT = "ai"
+    MESSAGES = "messages"
+    def initialize_session_state():
+        if MESSAGES not in st.session_state:
+            st.session_state[MESSAGES] = [Message(actor=ASSISTANT, payload="Hi!How can I help you?")]
+        if "llm_chain" not in st.session_state:
+            st.session_state["llm_chain"] = get_conversational_chain()
+
+    def get_llm_chain_from_session():
+        return st.session_state["llm_chain"]
+
+    initialize_session_state()
+
+    msg: Message
+    for msg in st.session_state[MESSAGES]:
+        st.chat_message(msg.actor).write(msg.payload)
+
+    prompt = st.chat_input("Enter a prompt here")
+
+    if prompt:
+        docs = db.similarity_search(prompt, k=5)
+        st.session_state[MESSAGES].append(Message(actor=USER, payload=prompt))
+        st.chat_message(USER).write(prompt)
+
+        with st.spinner("Please wait.."):     
+            llm_chain = get_llm_chain_from_session()
+            response = llm_chain({"context": docs, "question": prompt})['text']
+
+            st.session_state[MESSAGES].append(Message(actor=ASSISTANT, payload=response))
+            st.chat_message(ASSISTANT).write(response)

doc_loading.py

@@ -0,0 +1,24 @@
+import requests
+from bs4 import BeautifulSoup
+import re
+import os
+from PyPDF2 import PdfReader
+
+def read_pdf_text(pdf_file):
+    # Reading the pdf
+    pdf_reader = PdfReader(pdf_file)
+    all_text = ""
+    # make it limited. min(5, len(pages))
+    for page in pdf_reader.pages:
+        all_text += page.extract_text()
+    return all_text
+
+def parse_linkedin_pdf(pdf_text):
+    sections = re.split(r'\n(?=\b(?:Experience|Contact|Education|Top Skills|Languages|Honors-Awards)\b)', pdf_text)
+    parsed_data = {}
+    for section in sections:
+        lines = section.split('\n')
+        section_name = lines[0]
+        section_text = '\n'.join(lines[1:])
+        parsed_data[section_name] = section_text
+    return parsed_data

llm_functions.py

@@ -0,0 +1,31 @@
+from langchain_openai import ChatOpenAI
+from langchain.prompts import PromptTemplate
+from langchain.chains.question_answering import load_qa_chain
+from langchain.chains import LLMChain
+from langchain.memory import ConversationBufferMemory
+from langchain.chains import RetrievalQA
+
+import streamlit as st
+secrets = st.secrets["openai"]
+
+def get_conversational_chain():
+    prompt_template = """You are an expert and polite HR.
+    In the context, a candidate's resume will be provided to you. Given a question the hiring manager wants to know about the candidate, i want you to give the answer with the most precision. Feel free to answer in sentences or bullet points whatever you find suitable.
+    if there is some "\n" imagine things are writting in separate lines. make your move accordingly 
+    If the question has no answer present in the resume, 
+    feel free to say, "try ansking something else, this information is not available", don't provide the wrong answer no matter what is present in the question\n\n
+    Context:\n {context}?\n
+    Question: \n{question}\n
+
+    Answer:
+    """
+    model = ChatOpenAI(temperature=0.7, api_key=secrets["OPENAI_API_KEY"])
+    memory = ConversationBufferMemory(llm=model, input_key = 'question', memory_key="chat_history")
+
+    prompt = PromptTemplate(template=prompt_template, 
+                            input_variables=["context"])
+    chain = LLMChain(llm=model, prompt = prompt,
+                          memory=memory)
+
+    return chain
+

requirements.txt

@@ -0,0 +1,11 @@
+langchain
+openai
+langchain_openai
+PyPDF2
+python-dotenv
+requests
+beautifulsoup4==4.12.3
+requests==2.26.0
+streamlit
+counter
+faiss-cpu