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

# Load the trained model
model = pickle.load(open('model.sav', 'rb'))

# Set page config and style
st.set_page_config(page_title='Student Placement Prediction', page_icon=':mortar_board:')
st.markdown(
    """
    <style>
    .main {
        max-width: 800px;
        margin: 0 auto;
    }
    .title {
        text-align: center;
        font-size: 32px;
        margin-top: 20px;
        margin-bottom: 30px;
    }
    .header {
        font-size: 24px;
        margin-top: 20px;
        margin-bottom: 10px;
    }
    .subheader {
        font-size: 20px;
        margin-top: 10px;
        margin-bottom: 10px;
    }
    .prediction {
        text-align: center;
        font-size: 24px;
        font-weight: bold;
        margin-top: 30px;
    }
    </style>
    """,
    unsafe_allow_html=True
)

st.title('Student Placement Prediction')
st.markdown('<div class="main">', unsafe_allow_html=True)

st.sidebar.header('Student Data')

# Function to get user input
def get_user_input():
    gender = st.sidebar.selectbox('Gender', ['Male', 'Female'])
    ssc_p = st.sidebar.slider('SSC Percentage', 0.0, 100.0, 67.0)
    ssc_b = st.sidebar.selectbox('SSC Board', ['Central', 'Others'])
    hsc_p = st.sidebar.slider('HSC Percentage', 0.0, 100.0, 91.0)
    hsc_b = st.sidebar.selectbox('HSC Board', ['Central', 'Others'])
    hsc_s = st.sidebar.selectbox('HSC Stream', ['Science', 'Commerce', 'Arts'])
    degree_p = st.sidebar.slider('Degree Percentage', 0.0, 100.0, 58.0)
    degree_t = st.sidebar.selectbox('Degree Field', ['Sci&Tech', 'Comm&Mgmt', 'Others'])
    workex = st.sidebar.selectbox('Work Experience', ['No', 'Yes'])
    etest_p = st.sidebar.slider('Employability Test Percentage', 0.0, 100.0, 55.0)
    specialisation = st.sidebar.selectbox('MBA Specialization', ['Mkt&HR', 'Mkt&Fin'])
    mba_p = st.sidebar.slider('MBA Percentage', 0.0, 100.0, 58.8)

    user_data = {
        'gender': 0 if gender == 'Male' else 1,
        'ssc_p': ssc_p,
        'ssc_b': 0 if ssc_b == 'Central' else 1,
        'hsc_p': hsc_p,
        'hsc_b': 0 if hsc_b == 'Central' else 1,
        'hsc_s': 0 if hsc_s == 'Science' else 1 if hsc_s == 'Commerce' else 2,
        'degree_p': degree_p,
        'degree_t': 0 if degree_t == 'Sci&Tech' else 1 if degree_t == 'Comm&Mgmt' else 2,
        'workex': 0 if workex == 'No' else 1,
        'etest_p': etest_p,
        'specialisation': 0 if specialisation == 'Mkt&HR' else 1,
        'mba_p': mba_p
    }

    user_data_df = pd.DataFrame(user_data, index=[0])
    return user_data_df

user_data_df = get_user_input()

st.markdown('<div class="header">Student Data</div>', unsafe_allow_html=True)
st.write(user_data_df)

if st.button('Predict'):
    prediction = model.predict(user_data_df)
    placement = "Placed" if prediction[0] == 1 else "Not Placed"
    prediction_text = f"Placement Prediction: <span style='color: green;'>{placement}</span>" if placement == "Placed" else f"Placement Prediction: <span style='color: red;'>{placement}</span>"
    st.markdown(f"<div class='prediction'>{prediction_text}</div>", unsafe_allow_html=True)

st.markdown('</div>', unsafe_allow_html=True)