# Streamlit uygulaması

import streamlit as st
import pandas as pd
import joblib
from catboost import CatBoostClassifier

st.set_page_config(
    page_title="Academic Success Classification",
    page_icon="🎓",
    layout="centered"
)

st.title("🎓 Academic Success Classification / Akademik Başarı Sınıflandırma")

st.write("""
Bu uygulama; öğrencinin demografik bilgileri, akademik performansı ve ekonomik göstergeleri kullanarak
**Mezun Olma (Graduate)**, **Eğitime Devam Etme (Enrolled)** veya
**Okulu Bırakma Riski (Dropout)** durumlarından hangisine daha yakın olduğunu
makine öğrenmesi modeli ile tahmin eder.

This application uses a student's demographic information, academic performance,
and socioeconomic indicators to predict whether the student is most likely to
**Graduate**, **Remain Enrolled**, or **Drop Out**.
""")

model = CatBoostClassifier()
model.load_model("src/academic_success_model.cbm")

feature_columns = joblib.load("src/feature_columns.pkl")
reverse_target_map = joblib.load("src/reverse_target_map.pkl")

marital_status = st.number_input("Marital Status / Medeni Durum", min_value=1, max_value=6, value=1)
application_mode = st.number_input("Application Mode / Başvuru Şekli", min_value=1, max_value=60, value=1)
application_order = st.number_input("Application Order / Başvuru Sırası", min_value=0, max_value=10, value=1)
course = st.number_input("Course / Program", min_value=1, max_value=9999, value=9500)
daytime_evening_attendance = st.number_input("Daytime/Evening Attendance / Eğitim Zamanı", min_value=0, max_value=1, value=1)
previous_qualification = st.number_input("Previous Qualification / Önceki Eğitim", min_value=1, max_value=50, value=1)
previous_qualification_grade = st.number_input("Previous Qualification Grade / Önceki Eğitim Notu", min_value=0.0, max_value=200.0, value=130.0)
nationality = st.number_input("Nationality / Uyruk", min_value=1, max_value=200, value=1)
mothers_qualification = st.number_input("Mother's Qualification / Anne Eğitim Durumu", min_value=1, max_value=50, value=1)
fathers_qualification = st.number_input("Father's Qualification / Baba Eğitim Durumu", min_value=1, max_value=50, value=1)
mothers_occupation = st.number_input("Mother's Occupation / Anne Mesleği", min_value=0, max_value=200, value=1)
fathers_occupation = st.number_input("Father's Occupation / Baba Mesleği", min_value=0, max_value=200, value=1)

admission_grade = st.number_input("Admission Grade / Kabul Notu", min_value=0.0, max_value=200.0, value=130.0)
displaced = st.selectbox("Displaced / Yerinden Edilmiş", [0, 1])
educational_special_needs = st.selectbox("Educational Special Needs / Özel Eğitim İhtiyacı", [0, 1])
debtor = st.selectbox("Debtor / Borçlu", [0, 1])
tuition_fees_up_to_date = st.selectbox("Tuition Fees Up to Date / Harç Ödemesi Güncel", [0, 1])
gender = st.selectbox("Gender / Cinsiyet", [0, 1])
scholarship_holder = st.selectbox("Scholarship Holder / Burslu", [0, 1])
age_at_enrollment = st.number_input("Age at Enrollment / Kayıt Yaşı", min_value=15, max_value=80, value=20)
international = st.selectbox("International / Uluslararası Öğrenci", [0, 1])

cu1_credited = st.number_input("1st Sem Credited Units", min_value=0, max_value=30, value=0)
cu1_enrolled = st.number_input("1st Sem Enrolled Units", min_value=0, max_value=30, value=6)
cu1_evaluations = st.number_input("1st Sem Evaluations", min_value=0, max_value=50, value=6)
cu1_approved = st.number_input("1st Sem Approved Units", min_value=0, max_value=30, value=5)
cu1_grade = st.number_input("1st Sem Grade", min_value=0.0, max_value=20.0, value=12.0)
cu1_without_evaluations = st.number_input("1st Sem Without Evaluations", min_value=0, max_value=30, value=0)

cu2_credited = st.number_input("2nd Sem Credited Units", min_value=0, max_value=30, value=0)
cu2_enrolled = st.number_input("2nd Sem Enrolled Units", min_value=0, max_value=30, value=6)
cu2_evaluations = st.number_input("2nd Sem Evaluations", min_value=0, max_value=50, value=6)
cu2_approved = st.number_input("2nd Sem Approved Units", min_value=0, max_value=30, value=5)
cu2_grade = st.number_input("2nd Sem Grade", min_value=0.0, max_value=20.0, value=12.0)
cu2_without_evaluations = st.number_input("2nd Sem Without Evaluations", min_value=0, max_value=30, value=0)

unemployment_rate = st.number_input("Unemployment Rate / İşsizlik Oranı", min_value=0.0, max_value=30.0, value=10.0)
inflation_rate = st.number_input("Inflation Rate / Enflasyon Oranı", min_value=-5.0, max_value=20.0, value=1.0)
gdp = st.number_input("GDP / GSYH", min_value=-10.0, max_value=10.0, value=1.0)

academic_success = cu1_grade + cu2_grade
total_approved = cu1_approved + cu2_approved
approval_rate = total_approved / (cu1_enrolled + cu2_enrolled + 1)
performance_change = cu2_grade - cu1_grade
evaluation_efficiency = total_approved / (cu1_evaluations + cu2_evaluations + 1)

input_df = pd.DataFrame({
    "Marital status": [marital_status],
    "Application mode": [application_mode],
    "Application order": [application_order],
    "Course": [course],
    "Daytime/evening attendance": [daytime_evening_attendance],
    "Previous qualification": [previous_qualification],
    "Previous qualification (grade)": [previous_qualification_grade],
    "Nacionality": [nationality],
    "Mother's qualification": [mothers_qualification],
    "Father's qualification": [fathers_qualification],
    "Mother's occupation": [mothers_occupation],
    "Father's occupation": [fathers_occupation],
    "Admission grade": [admission_grade],
    "Displaced": [displaced],
    "Educational special needs": [educational_special_needs],
    "Debtor": [debtor],
    "Tuition fees up to date": [tuition_fees_up_to_date],
    "Gender": [gender],
    "Scholarship holder": [scholarship_holder],
    "Age at enrollment": [age_at_enrollment],
    "International": [international],
    "Curricular units 1st sem (credited)": [cu1_credited],
    "Curricular units 1st sem (enrolled)": [cu1_enrolled],
    "Curricular units 1st sem (evaluations)": [cu1_evaluations],
    "Curricular units 1st sem (approved)": [cu1_approved],
    "Curricular units 1st sem (grade)": [cu1_grade],
    "Curricular units 1st sem (without evaluations)": [cu1_without_evaluations],
    "Curricular units 2nd sem (credited)": [cu2_credited],
    "Curricular units 2nd sem (enrolled)": [cu2_enrolled],
    "Curricular units 2nd sem (evaluations)": [cu2_evaluations],
    "Curricular units 2nd sem (approved)": [cu2_approved],
    "Curricular units 2nd sem (grade)": [cu2_grade],
    "Curricular units 2nd sem (without evaluations)": [cu2_without_evaluations],
    "Unemployment rate": [unemployment_rate],
    "Inflation rate": [inflation_rate],
    "GDP": [gdp],
    "academic_success": [academic_success],
    "total_approved": [total_approved],
    "approval_rate": [approval_rate],
    "performance_change": [performance_change],
    "evaluation_efficiency": [evaluation_efficiency]
})

input_df = input_df[feature_columns]

if st.button("Classify Academic Outcome / Akademik Durumu Sınıflandır"):

    prediction_code = int(model.predict(input_df)[0][0])
    prediction = reverse_target_map[prediction_code]

    prediction_tr = {
        "Graduate": "Mezun Olur",
        "Enrolled": "Eğitimine Devam Eder",
        "Dropout": "Okulu Bırakma Riski"
    }[prediction]

    if prediction == "Graduate":
        st.success(f"🎓 Predicted Academic Outcome: {prediction}")
        st.success(f"🎓 Tahmini Akademik Durum: {prediction_tr}")

    elif prediction == "Enrolled":
        st.info(f"📚 Predicted Academic Outcome: {prediction}")
        st.info(f"📚 Tahmini Akademik Durum: {prediction_tr}")

    else:
        st.error(f"⚠️ Predicted Academic Outcome: {prediction}")
        st.error(f"⚠️ Tahmini Akademik Durum: {prediction_tr}")