recommender.py

import pandas as pd

def score_skills(user_skills, skills_df):
    return int((len(user_skills) / len(skills_df)) * 100)

def classify_field(text, field_labels):
    # Replace with a transformer model if needed
    for field in field_labels:
        if field.lower() in text.lower():
            return field
    return "General"

def recommend_countries(user_skills, countries_df):
    return countries_df[countries_df["Skill"].isin(user_skills)].drop_duplicates()

def recommend_certifications(user_skills, cert_df):
    return cert_df[cert_df["Skill"].isin(user_skills)].drop_duplicates()

def recommend_education(background, edu_tech_df, edu_non_tech_df):
    return edu_tech_df if background == "technical" else edu_non_tech_df

def generate_roadmap(cv_data, field, score, country_jobs, certs, scholarships):
    roadmap = f"""
    ## 📍 Your Personalized Career Roadmap

    - **Field**: {field}
    - **Skill Score**: {score}/100
    - **Years of Experience**: {cv_data['years_experience']}
    
    ### 🌍 Country-Based Job Opportunities:
    {country_jobs.to_markdown(index=False) if not country_jobs.empty else 'No matching jobs found.'}
    
    ### 🎓 Recommended Certifications:
    {certs.to_markdown(index=False) if not certs.empty else 'None found.'}
    
    ### 🎓 Suggested Scholarships:
    {scholarships.to_markdown(index=False) if not scholarships.empty else 'None found.'}

    ### 🧭 Next Steps:
    - Improve missing skills.
    - Pursue listed certifications.
    - Apply to the recommended countries.
    - Consider higher education if needed.

    """
    return roadmap