streamlit_app.py

import streamlit as st
import pandas as pd
import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import re
import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
import string
import json
import os
from utils import load_job_data, preprocess_text, get_top_skills, get_job_recommendations

# Download NLTK resources
try:
    nltk.download('punkt')
    nltk.download('stopwords')
except:
    pass  # In case downloads fail, continue with what's available

# Page configuration
st.set_page_config(
    page_title="Job Title Recommender",
    page_icon="💼",
    layout="wide"
)

# Custom CSS
st.markdown("""
<style>
    .main-header {
        font-size: 2.5rem;
        font-weight: bold;
        color: #0A66C2;
        text-align: center;
        margin-bottom: 1rem;
    }
    .sub-header {
        font-size: 1.2rem;
        color: #666;
        text-align: center;
        margin-bottom: 2rem;
    }
    .footer {
        text-align: center;
        margin-top: 2rem;
        color: #666;
        font-size: 0.9rem;
    }
    .recommendation-card {
        border: 1px solid #ddd;
        border-radius: 8px;
        padding: 1rem;
        margin-bottom: 1rem;
        background-color: #f9f9f9;
    }
    .skill-tag {
        display: inline-block;
        background-color: #0A66C2;
        color: white;
        padding: 0.3rem 0.8rem;
        border-radius: 15px;
        margin: 0.2rem;
        font-size: 0.9rem;
    }
</style>
""", unsafe_allow_html=True)

# Initialize session state
if 'job_data' not in st.session_state:
    st.session_state.job_data = load_job_data()
if 'vectorizer' not in st.session_state:
    st.session_state.vectorizer = None
if 'job_vectors' not in st.session_state:
    st.session_state.job_vectors = None

# Header
st.markdown('<div class="main-header">💼 Job Title Recommender</div>', unsafe_allow_html=True)
st.markdown('<div class="sub-header">Find the perfect job titles based on your LinkedIn profile or resume</div>', unsafe_allow_html=True)

# Built with anycoder
st.markdown('<div style="text-align: center; margin-bottom: 1rem;">Built with <a href="https://huggingface.co/spaces/akhaliq/anycoder" target="_blank">anycoder</a></div>', unsafe_allow_html=True)

# Main content
col1, col2 = st.columns([1, 1])

with col1:
    st.header("📝 Input Your Profile")

    # Input method selection
    input_method = st.radio(
        "Choose input method:",
        ["Text Input", "File Upload"],
        horizontal=True
    )

    if input_method == "Text Input":
        profile_text = st.text_area(
            "Paste your LinkedIn profile or resume text here:",
            height=300,
            placeholder="Include your work experience, skills, education, and any other relevant information..."
        )
    else:
        uploaded_file = st.file_uploader(
            "Upload your resume (PDF or TXT):",
            type=["pdf", "txt"],
            help="Upload your resume file to analyze"
        )

        if uploaded_file:
            try:
                if uploaded_file.type == "application/pdf":
                    # For PDF files (would need additional libraries)
                    st.warning("PDF parsing requires additional libraries. Please use text input or upload a .txt file.")
                    profile_text = ""
                else:
                    # For text files
                    profile_text = uploaded_file.read().decode("utf-8")
                    st.success("File uploaded successfully!")
            except Exception as e:
                st.error(f"Error reading file: {e}")
                profile_text = ""

    # Skills input
    st.subheader("Key Skills")
    skills_input = st.text_input(
        "Enter your key skills (comma separated):",
        placeholder="e.g., Python, Machine Learning, Project Management"
    )

    # Experience level
    experience_level = st.select_slider(
        "Years of Experience:",
        options=["0-2 years", "3-5 years", "6-10 years", "10+ years"],
        value="3-5 years"
    )

    # Education level
    education_level = st.selectbox(
        "Highest Education Level:",
        ["High School", "Associate Degree", "Bachelor's Degree", "Master's Degree", "PhD"]
    )

    # Analyze button
    analyze_button = st.button("🔍 Analyze & Get Recommendations", type="primary")

with col2:
    st.header("🎯 Your Recommendations")

    if analyze_button and profile_text:
        with st.spinner("Analyzing your profile..."):
            try:
                # Preprocess the input text
                processed_text = preprocess_text(profile_text)

                # Extract skills from input
                input_skills = [skill.strip() for skill in skills_input.split(",")] if skills_input else []
                extracted_skills = get_top_skills(processed_text, top_n=10)

                # Combine all skills
                all_skills = list(set(input_skills + extracted_skills))

                # Create user profile vector
                if st.session_state.vectorizer is None or st.session_state.job_vectors is None:
                    # Initialize vectorizer and job vectors if not already done
                    job_descriptions = st.session_state.job_data['description'].tolist()
                    st.session_state.vectorizer = TfidfVectorizer(stop_words='english', max_features=5000)
                    st.session_state.job_vectors = st.session_state.vectorizer.fit_transform(job_descriptions)

                # Transform user profile
                user_vector = st.session_state.vectorizer.transform([processed_text])

                # Get recommendations
                recommendations = get_job_recommendations(
                    user_vector,
                    st.session_state.job_vectors,
                    st.session_state.job_data,
                    all_skills,
                    experience_level,
                    education_level,
                    top_n=5
                )

                if recommendations.empty:
                    st.warning("No suitable job titles found. Try adding more details to your profile.")
                else:
                    st.success(f"Found {len(recommendations)} suitable job titles for you!")

                    for idx, row in recommendations.iterrows():
                        with st.expander(f"🏆 {row['job_title']} (Match: {row['match_score']:.1%})"):
                            st.markdown(f"**Industry:** {row['industry']}")
                            st.markdown(f"**Experience Level:** {row['experience_level']}")
                            st.markdown(f"**Education Requirement:** {row['education_requirement']}")

                            # Display skills
                            st.markdown("**Key Skills:**")
                            job_skills = [skill.strip() for skill in row['required_skills'].split(",")]
                            for skill in job_skills[:5]:  # Show top 5 skills
                                st.markdown(f'<span class="skill-tag">{skill}</span>', unsafe_allow_html=True)

                            st.markdown(f"**Average Salary:** {row['avg_salary']}")

                            if st.button(f"Learn more about {row['job_title']}", key=f"learn_{idx}"):
                                st.info(f"This would typically link to more information about the {row['job_title']} role.")

                    # Show skills analysis
                    st.subheader("Your Skills Analysis")
                    if all_skills:
                        st.markdown("**Detected Skills:**")
                        for skill in all_skills[:10]:  # Show top 10 skills
                            st.markdown(f'<span class="skill-tag">{skill}</span>', unsafe_allow_html=True)
                    else:
                        st.info("No specific skills detected. Consider adding more details to your profile.")

            except Exception as e:
                st.error(f"An error occurred during analysis: {e}")
    elif analyze_button and not profile_text:
        st.warning("Please enter your profile information or upload a file.")
    else:
        st.info("Enter your LinkedIn profile or resume information and click 'Analyze' to get job recommendations.")

# Footer
st.markdown("""
<div class="footer">
    <p>This tool uses natural language processing to match your skills and experience with suitable job titles.</p>
    <p>For best results, provide detailed information about your work experience, skills, and education.</p>
</div>
""", unsafe_allow_html=True)