Upload 4 files

Dockerfile

@@ -0,0 +1,25 @@
+FROM python:3.12-slim
+
+ENV DEBIAN_FRONTEND=noninteractive
+
+# Install essential packages
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    libpq-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+# Set working directory
+WORKDIR /home/user/app
+
+# Copy your code
+COPY . .
+
+# Install Python dependencies
+RUN pip install --no-cache-dir --upgrade pip
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Expose port for Streamlit
+EXPOSE 8501
+
+# Run app
+CMD ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0"]

README.md

@@ -0,0 +1,32 @@
+---
+title: AI Resume Screening App
+emoji: 🔍
+colorFrom: blue
+colorTo: green
+sdk: streamlit
+sdk_version: 1.35.0
+app_file: app.py
+pinned: false
+---
+
+# 🔍 AI Resume Screening App
+
+This project is an AI-powered resume screening tool built with Python, Streamlit, spaCy, and scikit-learn. It processes PDF and DOCX resumes, extracts skills and experience, and computes a match score against a job description.
+
+## 🚀 Features
+- Extracts text from PDF and DOCX resumes.
+- Identifies user-defined skills and estimates years of experience.
+- Computes a cosine similarity score between resume and job description.
+- Streamlit UI for deployment on Hugging Face Spaces.
+- Runs on CPU.
+
+## 📦 Setup on Hugging Face Spaces
+1. Upload resumes (PDF/DOCX) to `/data/resumes` via the Files tab (optional, as Streamlit handles uploads).
+2. Access the Streamlit interface and enter a job description and required skills (comma-separated).
+3. Upload resumes and view screening results.
+4. Check `/data/resumes` for persistent storage of uploaded files.
+
+## 📋 Requirements
+Install dependencies using:
+```bash
+pip install -r requirements.txt

requirements.txt

@@ -0,0 +1,16 @@
+# Core ML
+numpy
+scipy
+pandas
+scikit-learn
+
+# Text Processing
+spacy
+pdfplumber
+docx2txt
+
+# Web Framework
+streamlit==1.35.0
+
+# spaCy English Model
+en-core-web-sm @ https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.7.1/en_core_web_sm-3.7.1-py3-none-any.whl

resume.py

@@ -0,0 +1,107 @@
+import streamlit as st
+import pdfplumber
+import docx2txt
+import spacy
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+import os
+
+# Load the English NLP model from spaCy
+@st.cache_resource
+def load_spacy_model():
+    return spacy.load('en_core_web_sm')
+
+nlp = load_spacy_model()
+
+# Function to extract text from a PDF file
+def extract_text_from_pdf(pdf_file):
+    text = ''
+    with pdfplumber.open(pdf_file) as pdf:
+        for page in pdf.pagesSNS
+            page_text = page.extract_text()
+            if page_text:
+                text += page_text
+    return text
+
+# Function to extract text from a DOCX file
+def extract_text_from_docx(docx_file):
+    return docx2txt.process(docx_file)
+
+# Function to extract user-defined skills from resume text
+def extract_skills(text, user_skills):
+    text = text.lower()
+    extracted = [skill.strip().lower() for skill in user_skills if skill.strip().lower() in text]
+    return list(set(extracted))  # remove duplicates
+
+# Function to estimate years of experience from dates mentioned
+def extract_experience(text):
+    doc = nlp(text)
+    years = []
+    for ent in doc.ents:
+        if ent.label_ == 'DATE':
+            try:
+                if 'year' in ent.text.lower():
+                    num = int(ent.text.split()[0])
+                    years.append(num)
+            except:
+                continue
+    return max(years, default=0)
+
+# Function to compute a similarity score between resume and job description
+def match_score(resume_text, job_description):
+    documents = [resume_text, job_description]
+    tfidf = TfidfVectorizer(stop_words='english')
+    tfidf_matrix = tfidf.fit_transform(documents)
+    score = cosine_similarity(tfidf_matrix[0:1], tfidf_matrix[1:2])
+    return round(float(score[0][0]) * 100, 2)
+
+# -------- Streamlit Frontend Starts Here -------- #
+
+st.title("🔍 AI Resume Screening App")
+
+# Text area for job description
+job_description = st.text_area("📄 Paste the Job Description Below:", height=200)
+
+# Text input for skills (comma-separated)
+skills_input = st.text_input("🛠️ Enter Required Skills (comma-separated):", placeholder="e.g., Python, SQL, Machine Learning")
+
+# File uploader for multiple resumes
+uploaded_files = st.file_uploader("📂 Upload Resume Files (PDF/DOCX)", type=['pdf', 'docx'], accept_multiple_files=True)
+
+# Main logic to process resumes
+if uploaded_files and job_description and skills_input:
+    # Parse user-entered skills
+    user_skills = [skill.strip() for skill in skills_input.split(',') if skill.strip()]
+    
+    if not user_skills:
+        st.warning("⚠️ Please enter at least one skill.")
+    else:
+        st.markdown("### 🔎 Screening Results")
+
+        # Save uploaded files to /data for persistent storage
+        os.makedirs('/data/resumes', exist_ok=True)
+        for resume in uploaded_files:
+            resume_path = os.path.join('/data/resumes', resume.name)
+            with open(resume_path, 'wb') as f:
+                f.write(resume.read())
+
+            # Extract text
+            if resume.name.endswith('.pdf'):
+                resume_text = extract_text_from_pdf(resume_path)
+            elif resume.name.endswith('.docx'):
+                resume_text = extract_text_from_docx(resume_path)
+            else:
+                st.warning(f"Unsupported file type: {resume.name}")
+                continue
+
+            # Extract information
+            skills = extract_skills(resume_text, user_skills)
+            experience = extract_experience(resume_text)
+            score = match_score(resume_text, job_description)
+
+            # Display results
+            st.subheader(f"👤 Candidate: {resume.name}")
+            st.write(f"✅ **Skills Matched**: {', '.join(skills) if skills else 'None'}")
+            st.write(f"🧠 **Estimated Experience**: {experience} year(s)")
+            st.write(f"📊 **Match Score**: {score}%")
+            st.markdown("---")