import streamlit as st
import pickle
import re
import docx
import PyPDF2
from sklearn.metrics.pairwise import cosine_similarity
# 1. CONFIG
st.set_page_config(
page_title="AI Resume Screening",
layout="wide",
initial_sidebar_state="collapsed"
)
# Custom CSS for styling
st.markdown("""
""", unsafe_allow_html=True)
# 2. LOAD RESOURCES
@st.cache_resource
def load_resources():
try:
clf = pickle.load(open('clf.pkl', 'rb'))
tfidf = pickle.load(open('tfidf.pkl', 'rb'))
le = pickle.load(open('encoder.pkl', 'rb'))
ats = pickle.load(open('ats_scorer.pkl', 'rb'))
prototypes = pickle.load(open('prototypes.pkl', 'rb'))
return clf, tfidf, le, ats, prototypes
except FileNotFoundError:
return None, None, None, None, None
clf, tfidf, le, ats_model, prototypes = load_resources()
# 3. UTILS
def clean_text(txt):
txt = re.sub(r'http\S+\s', ' ', txt)
txt = re.sub(r'[^\w\s]', ' ', txt)
return txt.lower()
def extract_text(file):
try:
if file.name.endswith('.pdf'):
reader = PyPDF2.PdfReader(file)
return " ".join([page.extract_text() for page in reader.pages])
elif file.name.endswith('.docx'):
doc = docx.Document(file)
return " ".join([p.text for p in doc.paragraphs])
elif file.name.endswith('.txt'):
return file.read().decode('utf-8')
except:
return ""
def calculate_scores(text, category):
if category not in prototypes:
return 0, 0, 0
master_profile = prototypes[category]
cleaned_resume = clean_text(text)
# Cosine Similarity
vecs = tfidf.transform([cleaned_resume, master_profile])
cosine_sim = cosine_similarity(vecs[0], vecs[1])[0][0]
# Keyword Match
res_tokens = set(cleaned_resume.split())
mp_tokens = set(master_profile.split())
keyword_match = len(res_tokens.intersection(mp_tokens)) / len(mp_tokens) if mp_tokens else 0
# AI Prediction
try:
ml_score = ats_model.predict([[cosine_sim, keyword_match]])[0]
except:
ml_score = 0
# Fallback Logic
if ml_score < 10:
final_score = cosine_sim * 100
else:
final_score = ml_score
if final_score < 1:
final_score *= 100
return round(final_score, 1), round(cosine_sim*100, 1), round(keyword_match*100, 1)
# 4. MAIN APP
def main():
# Header
st.markdown("🎯 AI Resume Screening
", unsafe_allow_html=True)
st.markdown("Powered by Machine Learning & Natural Language Processing
", unsafe_allow_html=True)
if not clf:
st.error("⚠️ Models missing! Run `train_model.py` then `train_ats_model.py`.")
st.stop()
# Upload section
st.markdown("
", unsafe_allow_html=True)
col1, col2, col3 = st.columns([1, 2, 1])
with col2:
file = st.file_uploader(
"📤 Upload Your Resume",
type=['pdf', 'docx', 'txt'],
help="Supported formats: PDF, DOCX, TXT"
)
if file:
# Custom loading animation
loading_placeholder = st.empty()
with loading_placeholder.container():
st.markdown("""
""", unsafe_allow_html=True)
# Extract and process
text = extract_text(file)
# Clear loading animation
loading_placeholder.empty()
if len(text) > 20:
clean = clean_text(text)
vec = tfidf.transform([clean])
cat_id = clf.predict(vec)[0]
category = le.inverse_transform([cat_id])[0]
ats_score, raw_sim, key_match = calculate_scores(text, category)
st.markdown("
", unsafe_allow_html=True)
# Category prediction
st.success(f"### 🎯 Predicted Role: **{category}**")
# Score badge
if ats_score >= 75:
badge_class = "score-high"
emoji = "🌟"
elif ats_score >= 50:
badge_class = "score-medium"
emoji = "⚡"
else:
badge_class = "score-low"
emoji = "💡"
st.markdown(f"{emoji} ATS Score: {ats_score}%
", unsafe_allow_html=True)
# Metrics
st.markdown("### 📊 Detailed Analysis")
col1, col2, col3 = st.columns(3)
with col1:
st.metric(
label="🤖 AI Score",
value=f"{ats_score}%",
delta="Primary Score"
)
with col2:
st.metric(
label="📝 Content Match",
value=f"{raw_sim}%",
delta="Similarity"
)
with col3:
st.metric(
label="🔑 Keywords",
value=f"{key_match}%",
delta="Overlap"
)
# Progress bar
st.markdown("#### Match Strength")
st.progress(min(ats_score/100, 1.0))
# Feedback
if ats_score > 75:
st.balloons()
st.info("🎉 Excellent match! Your resume aligns well with this role.")
elif ats_score >= 50:
st.info("✨ Good match! Consider adding more role-specific keywords to improve.")
else:
st.warning("💡 Low match. Try adding more relevant skills and experience keywords.")
# Extracted text
with st.expander("📄 View Extracted Text"):
st.text_area("Resume Content", text, height=300)
else:
st.warning("⚠️ Could not extract text. File might be an image or scan. Please use a text-based document.")
if __name__ == "__main__":
main()