Py.Cafe

import streamlit as st
import pandas as pd
import numpy as np

# Try importing packages, if not available, show error
try:
    from sklearn.ensemble import RandomForestClassifier
    from sklearn.preprocessing import StandardScaler, LabelEncoder
    import plotly.graph_objects as go
except ImportError:
    st.error("Required packages not installed. This app needs scikit-learn and plotly.")
    st.stop()

# Page config
st.set_page_config(
    page_title="ni - Neuron Inputs Career Finder",
    page_icon="🧠",
    layout="wide"
)

# Custom CSS
st.markdown("""
    <style>
    .main {
        padding: 2rem;
    }
    .stButton>button {
        width: 100%;
        background-color: #4CAF50;
        color: white;
        height: 3em;
        border-radius: 10px;
        font-size: 18px;
    }
    </style>
    """, unsafe_allow_html=True)

# Initialize session state
if 'model_trained' not in st.session_state:
    st.session_state.model_trained = False

@st.cache_resource
def generate_training_data():
    """Generate diverse synthetic career dataset"""
    np.random.seed(42)
    
    # 40+ diverse career paths for Indian students
    careers = [
        # Tech & Engineering
        'Software Engineer', 'Data Scientist', 'AI/ML Engineer', 'Cybersecurity Specialist',
        'Web Developer', 'Mobile App Developer', 'Cloud Architect', 'DevOps Engineer',
        # Medical & Healthcare
        'Doctor (MBBS)', 'Dentist', 'Pharmacist', 'Physiotherapist', 'Nursing', 'Medical Research',
        # Business & Finance
        'Chartered Accountant', 'Investment Banker', 'Business Analyst', 'Financial Advisor',
        'Entrepreneur', 'Marketing Manager', 'HR Manager', 'Management Consultant',
        # Creative & Design
        'Graphic Designer', 'UI/UX Designer', 'Content Writer', 'Filmmaker', 'Animator',
        'Fashion Designer', 'Photographer', 'Video Editor',
        # Traditional Professions
        'Civil Engineer', 'Mechanical Engineer', 'Electrical Engineer', 'Architect',
        # Social & Legal
        'Lawyer', 'Teacher/Professor', 'Psychologist', 'Social Worker', 'Journalist',
        # Other
        'Chef', 'Interior Designer', 'Environmental Scientist', 'Biotechnologist'
    ]
    
    career_profiles = {
        'Software Engineer': {
            'math': (85, 8), 'science': (82, 8), 'english': (70, 12), 'social': (65, 12),
            'openness': (8, 1), 'conscientiousness': (7, 1.5), 'extraversion': (5, 2),
            'agreeableness': (6, 1.5), 'neuroticism': (5, 2),
            'technical': (9, 1), 'creative': (5, 2), 'business': (4, 2), 'social_service': (4, 2),
            'coding': (8, 1.5), 'communication': (6, 2), 'analysis': (8, 1.5), 'leadership': (5, 2)
        },
        'Data Scientist': {
            'math': (90, 6), 'science': (85, 8), 'english': (72, 10), 'social': (68, 12),
            'openness': (9, 1), 'conscientiousness': (8, 1), 'extraversion': (5, 2),
            'agreeableness': (6, 1.5), 'neuroticism': (4, 2),
            'technical': (9, 1), 'creative': (6, 2), 'business': (6, 2), 'social_service': (4, 2),
            'coding': (9, 1), 'communication': (7, 1.5), 'analysis': (9, 1), 'leadership': (6, 2)
        },
        'Doctor (MBBS)': {
            'math': (75, 10), 'science': (92, 6), 'english': (70, 12), 'social': (68, 12),
            'openness': (7, 1.5), 'conscientiousness': (9, 1), 'extraversion': (6, 2),
            'agreeableness': (8, 1.5), 'neuroticism': (4, 2),
            'technical': (5, 2), 'creative': (4, 2), 'business': (4, 2), 'social_service': (9, 1),
            'coding': (3, 2), 'communication': (7, 1.5), 'analysis': (8, 1.5), 'leadership': (6, 2)
        },
        'Chartered Accountant': {
            'math': (88, 8), 'science': (70, 12), 'english': (75, 10), 'social': (72, 12),
            'openness': (6, 1.5), 'conscientiousness': (9, 1), 'extraversion': (5, 2),
            'agreeableness': (7, 1.5), 'neuroticism': (5, 2),
            'technical': (5, 2), 'creative': (3, 2), 'business': (9, 1), 'social_service': (4, 2),
            'coding': (4, 2), 'communication': (7, 2), 'analysis': (9, 1), 'leadership': (6, 2)
        },
        'Graphic Designer': {
            'math': (60, 15), 'science': (60, 15), 'english': (75, 10), 'social': (70, 12),
            'openness': (9, 1), 'conscientiousness': (7, 1.5), 'extraversion': (6, 2),
            'agreeableness': (7, 1.5), 'neuroticism': (5, 2),
            'technical': (6, 2), 'creative': (9, 1), 'business': (5, 2), 'social_service': (4, 2),
            'coding': (5, 2), 'communication': (7, 1.5), 'analysis': (6, 2), 'leadership': (5, 2)
        },
        'Content Writer': {
            'math': (60, 15), 'science': (60, 15), 'english': (90, 6), 'social': (80, 10),
            'openness': (9, 1), 'conscientiousness': (7, 1.5), 'extraversion': (6, 2),
            'agreeableness': (7, 1.5), 'neuroticism': (5, 2),
            'technical': (4, 2), 'creative': (9, 1), 'business': (5, 2), 'social_service': (5, 2),
            'coding': (3, 2), 'communication': (9, 1), 'analysis': (7, 1.5), 'leadership': (5, 2)
        },
        'Lawyer': {
            'math': (70, 12), 'science': (65, 15), 'english': (88, 8), 'social': (85, 8),
            'openness': (7, 1.5), 'conscientiousness': (9, 1), 'extraversion': (7, 1.5),
            'agreeableness': (6, 2), 'neuroticism': (4, 2),
            'technical': (4, 2), 'creative': (5, 2), 'business': (6, 2), 'social_service': (7, 1.5),
            'coding': (3, 2), 'communication': (9, 1), 'analysis': (9, 1), 'leadership': (8, 1.5)
        },
        'Entrepreneur': {
            'math': (75, 12), 'science': (70, 12), 'english': (78, 10), 'social': (75, 10),
            'openness': (9, 1), 'conscientiousness': (8, 1.5), 'extraversion': (8, 1.5),
            'agreeableness': (6, 2), 'neuroticism': (3, 2),
            'technical': (6, 2), 'creative': (7, 1.5), 'business': (9, 1), 'social_service': (5, 2),
            'coding': (5, 2), 'communication': (8, 1.5), 'analysis': (8, 1.5), 'leadership': (9, 1)
        },
        'Teacher/Professor': {
            'math': (75, 12), 'science': (75, 12), 'english': (82, 10), 'social': (78, 10),
            'openness': (8, 1.5), 'conscientiousness': (8, 1.5), 'extraversion': (7, 1.5),
            'agreeableness': (9, 1), 'neuroticism': (4, 2),
            'technical': (5, 2), 'creative': (6, 2), 'business': (4, 2), 'social_service': (9, 1),
            'coding': (4, 2), 'communication': (9, 1), 'analysis': (7, 1.5), 'leadership': (7, 1.5)
        },
        'Architect': {
            'math': (80, 10), 'science': (75, 12), 'english': (72, 12), 'social': (70, 12),
            'openness': (9, 1), 'conscientiousness': (8, 1.5), 'extraversion': (6, 2),
            'agreeableness': (7, 1.5), 'neuroticism': (5, 2),
            'technical': (7, 1.5), 'creative': (9, 1), 'business': (5, 2), 'social_service': (5, 2),
            'coding': (4, 2), 'communication': (7, 1.5), 'analysis': (8, 1.5), 'leadership': (6, 2)
        },
        'Psychologist': {
            'math': (65, 15), 'science': (70, 12), 'english': (80, 10), 'social': (85, 8),
            'openness': (9, 1), 'conscientiousness': (8, 1.5), 'extraversion': (6, 2),
            'agreeableness': (9, 1), 'neuroticism': (4, 2),
            'technical': (4, 2), 'creative': (6, 2), 'business': (4, 2), 'social_service': (9, 1),
            'coding': (3, 2), 'communication': (9, 1), 'analysis': (8, 1.5), 'leadership': (6, 2)
        },
        'Marketing Manager': {
            'math': (70, 12), 'science': (65, 15), 'english': (82, 10), 'social': (78, 10),
            'openness': (8, 1.5), 'conscientiousness': (7, 1.5), 'extraversion': (8, 1.5),
            'agreeableness': (7, 1.5), 'neuroticism': (4, 2),
            'technical': (5, 2), 'creative': (7, 1.5), 'business': (9, 1), 'social_service': (5, 2),
            'coding': (4, 2), 'communication': (9, 1), 'analysis': (7, 1.5), 'leadership': (8, 1.5)
        },
    }
    
    # Generate generic profiles for remaining careers
    for career in careers:
        if career not in career_profiles:
            career_profiles[career] = {
                'math': (70, 15), 'science': (70, 15), 'english': (70, 15), 'social': (70, 15),
                'openness': (6, 2), 'conscientiousness': (6, 2), 'extraversion': (6, 2),
                'agreeableness': (6, 2), 'neuroticism': (5, 2),
                'technical': (5, 2), 'creative': (5, 2), 'business': (5, 2), 'social_service': (5, 2),
                'coding': (5, 2), 'communication': (6, 2), 'analysis': (6, 2), 'leadership': (5, 2)
            }
    
    data = []
    samples_per_career = 80
    
    for career in careers:
        profile = career_profiles[career]
        for _ in range(samples_per_career):
            record = {
                'math_score': np.clip(np.random.normal(*profile['math']), 0, 100),
                'science_score': np.clip(np.random.normal(*profile['science']), 0, 100),
                'english_score': np.clip(np.random.normal(*profile['english']), 0, 100),
                'social_score': np.clip(np.random.normal(*profile['social']), 0, 100),
                'openness': np.clip(np.random.normal(*profile['openness']), 1, 10),
                'conscientiousness': np.clip(np.random.normal(*profile['conscientiousness']), 1, 10),
                'extraversion': np.clip(np.random.normal(*profile['extraversion']), 1, 10),
                'agreeableness': np.clip(np.random.normal(*profile['agreeableness']), 1, 10),
                'neuroticism': np.clip(np.random.normal(*profile['neuroticism']), 1, 10),
                'technical_interest': np.clip(np.random.normal(*profile['technical']), 1, 10),
                'creative_interest': np.clip(np.random.normal(*profile['creative']), 1, 10),
                'business_interest': np.clip(np.random.normal(*profile['business']), 1, 10),
                'social_interest': np.clip(np.random.normal(*profile['social_service']), 1, 10),
                'coding_skill': np.clip(np.random.normal(*profile['coding']), 1, 10),
                'communication': np.clip(np.random.normal(*profile['communication']), 1, 10),
                'analysis': np.clip(np.random.normal(*profile['analysis']), 1, 10),
                'leadership': np.clip(np.random.normal(*profile['leadership']), 1, 10),
                'career': career
            }
            data.append(record)
    
    return pd.DataFrame(data)

@st.cache_resource
def train_model():
    """Train the AI model"""
    df = generate_training_data()
    
    X = df.drop('career', axis=1)
    y = df['career']
    
    scaler = StandardScaler()
    label_encoder = LabelEncoder()
    
    X_scaled = scaler.fit_transform(X)
    y_encoded = label_encoder.fit_transform(y)
    
    model = RandomForestClassifier(n_estimators=200, max_depth=20, random_state=42)
    model.fit(X_scaled, y_encoded)
    
    return model, scaler, label_encoder, X.columns.tolist()

def predict_careers(user_data, model, scaler, label_encoder, feature_names):
    """Get career predictions"""
    user_df = pd.DataFrame([user_data])
    user_scaled = scaler.transform(user_df)
    
    probabilities = model.predict_proba(user_scaled)[0]
    top_indices = np.argsort(probabilities)[-7:][::-1]
    
    recommendations = []
    for idx in top_indices:
        career = label_encoder.inverse_transform([idx])[0]
        score = probabilities[idx] * 100
        
        # Get top contributing features
        feature_importances = model.feature_importances_
        user_values = user_df.values[0]
        weighted = user_values * feature_importances
        top_features_idx = np.argsort(weighted)[-3:][::-1]
        top_features = [feature_names[i].replace('_', ' ').title() for i in top_features_idx]
        
        recommendations.append({
            'career': career,
            'score': score,
            'strengths': top_features
        })
    
    return recommendations

# Main App
st.title("🧠 ni - Neuron Inputs")
st.subheader("AI-Powered Career Path Finder")

st.markdown("---")

# Sidebar
with st.sidebar:
    st.markdown("### 📖 About ni")
    st.info("""
    **ni** analyzes:
    - 📊 Academic Performance
    - 🧬 Personality (Big Five)
    - 💡 Interests & Skills
    
    Get recommendations from 40+ careers!
    """)
    
    user_type = st.selectbox(
        "I am a:",
        ["Class 10 Student", "Class 12 Student", "Undergraduate", "Graduate", "Working Professional"]
    )
    
    st.markdown("---")
    st.caption("Built by a startup founder 🚀")

# Train model
with st.spinner("🤖 Training AI Model..."):
    model, scaler, label_encoder, feature_names = train_model()

st.success("✅ Model Ready! Complete assessment below.")

# Assessment Form
st.markdown("## 📝 Your Assessment")

tab1, tab2, tab3, tab4 = st.tabs(["📚 Academic", "🧬 Personality", "💡 Interests", "🎯 Skills"])

with tab1:
    st.markdown("### Academic Scores")
    col1, col2 = st.columns(2)
    with col1:
        math_score = st.slider("Math (%)", 0, 100, 75)
        science_score = st.slider("Science (%)", 0, 100, 75)
    with col2:
        english_score = st.slider("English (%)", 0, 100, 75)
        social_score = st.slider("Social Studies (%)", 0, 100, 75)

with tab2:
    st.markdown("### Personality (1-10)")
    col1, col2 = st.columns(2)
    with col1:
        openness = st.slider("Openness", 1, 10, 6, help="Curiosity, creativity")
        conscientiousness = st.slider("Conscientiousness", 1, 10, 6, help="Organization, discipline")
        extraversion = st.slider("Extraversion", 1, 10, 6, help="Sociability")
    with col2:
        agreeableness = st.slider("Agreeableness", 1, 10, 6, help="Cooperation")
        neuroticism = st.slider("Emotional Stability", 1, 10, 6, help="Stress handling")

with tab3:
    st.markdown("### Interests (1-10)")
    col1, col2 = st.columns(2)
    with col1:
        technical_interest = st.slider("Technical/STEM", 1, 10, 5)
        creative_interest = st.slider("Creative/Arts", 1, 10, 5)
    with col2:
        business_interest = st.slider("Business", 1, 10, 5)
        social_interest = st.slider("Social Service", 1, 10, 5)

with tab4:
    st.markdown("### Skills (1-10)")
    col1, col2 = st.columns(2)
    with col1:
        coding_skill = st.slider("Coding", 1, 10, 5)
        communication = st.slider("Communication", 1, 10, 6)
    with col2:
        analysis = st.slider("Analysis", 1, 10, 6)
        leadership = st.slider("Leadership", 1, 10, 5)

st.markdown("---")

# Generate Button
if st.button("🚀 Get My Career Recommendations", type="primary"):
    
    user_data = {
        'math_score': math_score, 'science_score': science_score,
        'english_score': english_score, 'social_score': social_score,
        'openness': openness, 'conscientiousness': conscientiousness,
        'extraversion': extraversion, 'agreeableness': agreeableness,
        'neuroticism': neuroticism, 'technical_interest': technical_interest,
        'creative_interest': creative_interest, 'business_interest': business_interest,
        'social_interest': social_interest, 'coding_skill': coding_skill,
        'communication': communication, 'analysis': analysis,
        'leadership': leadership
    }
    
    with st.spinner("🔮 Analyzing..."):
        recommendations = predict_careers(user_data, model, scaler, label_encoder, feature_names)
    
    st.markdown("---")
    st.markdown("## 🎯 Your Top Career Matches")
    
    # Top 5 careers
    for i, rec in enumerate(recommendations[:5], 1):
        col1, col2 = st.columns([4, 1])
        with col1:
            st.markdown(f"### {i}. {rec['career']}")
            st.caption(f"💪 Strengths: {', '.join(rec['strengths'])}")
        with col2:
            st.metric("Match", f"{rec['score']:.0f}%")
        st.progress(rec['score'] / 100)
        st.markdown("")
    
    # More options
    with st.expander("📋 More Career Options"):
        for i, rec in enumerate(recommendations[5:], 6):
            st.write(f"**{i}. {rec['career']}** - {rec['score']:.1f}%")
    
    st.success("🎉 Done! Research these careers and explore learning paths.")
    
    # Simple viz
    st.markdown("### 📊 Your Profile")
    profile_scores = {
        'Academic': (math_score + science_score + english_score + social_score) / 4,
        'Technical': technical_interest * 10,
        'Creative': creative_interest * 10,
        'Business': business_interest * 10,
        'Social': social_interest * 10
    }
    st.bar_chart(profile_scores)

# Footer
st.markdown("---")
st.markdown("""
    <div style='text-align: center; color: gray;'>
        <p>🧠 <strong>ni - Neuron Inputs</strong> | Beta v1.0</p>
        <p style='font-size: 12px;'>Feedback? DM on LinkedIn!</p>
    </div>
""", unsafe_allow_html=True)