from flask import Flask, request, render_template
import pandas as pd
import joblib
import numpy as np

app = Flask(__name__)

# Load model, encoders, and optional scaler
model = joblib.load('placement_model.pkl')
le_placement = joblib.load('le_placement.pkl')
le_internship = joblib.load('le_internship.pkl')
# scaler = joblib.load('scaler.pkl')  # Uncomment if you used scaling

# Feature columns in exact order used during training
features = ['IQ', 'Prev_Sem_Result', 'CGPA', 'Academic_Performance',
            'Extra_Curricular_Score', 'Communication_Skills', 'Projects_Completed',
            'Internship_Encoded']

@app.route('/', methods=['GET', 'POST'])
def index():
    prediction = None
    if request.method == 'POST':
        try:
            # Collect inputs and convert to proper type
            IQ = float(request.form['IQ'])
            Prev_Sem_Result = float(request.form['Prev_Sem_Result'])
            CGPA = float(request.form['CGPA'])
            Academic_Performance = float(request.form['Academic_Performance'])
            Extra_Curricular_Score = float(request.form['Extra_Curricular_Score'])
            Communication_Skills = float(request.form['Communication_Skills'])
            Projects_Completed = int(request.form['Projects_Completed'])
            Internship_Experience = request.form['Internship_Experience'].strip()

            # Handle unknown internship category
            if Internship_Experience not in le_internship.classes_:
                # Assign most frequent category from training
                Internship_Experience = le_internship.classes_[0]

            internship_encoded = le_internship.transform([Internship_Experience])[0]

            # Prepare DataFrame in the same feature order as training
            X = pd.DataFrame([[IQ, Prev_Sem_Result, CGPA, Academic_Performance,
                               Extra_Curricular_Score, Communication_Skills,
                               Projects_Completed, internship_encoded]],
                             columns=features)

            # Optional: scale features if model expects it
            # X = pd.DataFrame(scaler.transform(X), columns=features)

            # Make prediction
            pred_encoded = model.predict(X)[0]
            pred_label = le_placement.inverse_transform([pred_encoded])[0]

            prediction = f'Predicted Placement: {pred_label}'

        except ValueError:
            prediction = "Invalid input! Please enter numeric values."

        except Exception as e:
            # Catch any unexpected error
            prediction = f"Error in prediction: {str(e)}"

    return render_template('index.html', prediction=prediction)

@app.route('/about')
def about():
    return render_template('about.html')

@app.route('/canva')
def canva():
    return render_template('canva.html')

if __name__ == '__main__':
    app.run(debug=True)