app.py

import streamlit as st
import pdfplumber
import torch
from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline

# Load Hugging Face model
model_name = "deepset/roberta-base-squad2"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
qa_pipeline = pipeline(
    "question-answering",
    model=model,
    tokenizer=tokenizer,
    truncation=True,
    max_length=1024
)

# Function to extract text from PDF
def extract_text_from_pdf(pdf_file):
    with pdfplumber.open(pdf_file) as pdf:
        text = "\n".join([page.extract_text() for page in pdf.pages if page.extract_text()])
        return text if text.strip() else "No text found in the document."

# Function to extract information using LLM
def extract_info(text, question):
    if not text or len(text.strip()) < 50:
        return "Insufficient text for analysis."
    if not question or len(question.strip()) == 0:
        return "No valid question provided."

    try:
        result = qa_pipeline({
            "context": text[:1024],
            "question": question
        })
        return result.get("answer", "No relevant information found.")
    except Exception as e:
        return f"Error processing request: {str(e)}"

# Function to analyze resume vs job description
def analyze_resume_vs_job(resume_text, job_desc):
    job_keywords = set(job_desc.lower().split())  # Extract keywords from JD
    resume_keywords = set(resume_text.lower().split())  # Extract keywords from resume

    matched_skills = job_keywords.intersection(resume_keywords)  # Common words
    missing_skills = job_keywords.difference(resume_keywords)  # Missing words

    return matched_skills, missing_skills

# Streamlit UI
st.title("📄 AI-Powered ATS Resume Parser")
st.write("Upload a resume to extract key details automatically!")

uploaded_file = st.file_uploader("Upload Resume (PDF or TXT)", type=["pdf", "txt"])

if uploaded_file is not None:
    file_extension = uploaded_file.name.split(".")[-1]
    if file_extension == "pdf":
        resume_text = extract_text_from_pdf(uploaded_file)
    else:
        resume_text = uploaded_file.read().decode("utf-8")

    st.subheader("📌 Extracted Resume Information")
    name = extract_info(resume_text, "What is the applicant's name?")
    job_title = extract_info(resume_text, "What is the applicant's job title?")
    skills = extract_info(resume_text, "What are the applicant's skills?")
    experience = extract_info(resume_text, "How many years of experience does the applicant have?")

    st.write(f"**🧑 Name:** {name}")
    st.write(f"**💼 Job Title:** {job_title}")
    st.write(f"**🛠 Skills:** {skills}")
    st.write(f"**📅 Experience:** {experience} years")

    # Job Matching (Enhanced)
    st.subheader("🔍 Match Resume with Job Description")
    job_desc = st.text_area("Paste the Job Description Here:")

    if st.button("Match Resume"):
        if not job_desc.strip():
            st.warning("⚠️ Please enter a Job Description before matching.")
        else:
            match_score = extract_info(resume_text, f"How well does this resume match the job description: {job_desc}")
            matched_skills, missing_skills = analyze_resume_vs_job(resume_text, job_desc)

            st.write(f"**✅ Match Score:** {match_score}")
            st.write(f"**✔ Matched Skills:** {', '.join(matched_skills) if matched_skills else 'None'}")
            st.write(f"**❌ Missing Skills:** {', '.join(missing_skills) if missing_skills else 'None'}")