app.py

from flask import Flask, render_template, request, redirect, url_for, flash
import os
from joblib import load
import numpy as np

app = Flask(__name__)

# Set the secret key for session management
app.secret_key = os.urandom(24)

# Configurations
UPLOAD_FOLDER = "uploads/"
app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
app.config['DEBUG'] = True  # Enable debugging during development
app.config['ENV'] = 'development'

# Load models
gia_model = load(r'H:\DEV PATEL\2025\AI_In_Diamond_Industry\Model\linear_regression_model_gia_price.joblib')
grade_model = load(r'H:\DEV PATEL\2025\AI_In_Diamond_Industry\Model\linear_regression_model_grade_price.joblib')
bygrade_model = load(r'H:\DEV PATEL\2025\AI_In_Diamond_Industry\Model\linear_regression_model_bygrade_price.joblib')
makable_model = load(r'H:\DEV PATEL\2025\AI_In_Diamond_Industry\Model\linear_regression_model_makable_price.joblib')

# Load label encoders for categorical features (renamed variable to avoid conflict with built-in list)
encoder_list = ['Tag', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol', 'EngSym', 'EngFlo', 'EngNts', 'EngMikly', 'EngLab']
loaded_label_encoder = {}
for val in encoder_list:
    encoder_path = f"H:/DEV PATEL/2025/AI_In_Diamond_Industry/Label_encoders/label_encoder_{val}.joblib"
    loaded_label_encoder[val] = load(encoder_path)

@app.route('/', methods=['GET'])
def home():
    return render_template('home.html')

@app.route('/predict', methods=['GET', 'POST'])
def predict():
    # Redirect GET requests back to the home page
    if request.method == 'GET':
        return redirect(url_for('home'))

    try:
        # Retrieve form data and strip any extra spaces ['Tag', 'EngCts', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol', 'EngSym', 'EngFlo', 'EngNts', 'EngMikly', 'EngLab', 'EngAmt',
        Tag = request.form.get('Tag', '').strip()                        
        # ICarat = request.form.get('ICarat', '').strip()
        # MkblCarat = request.form.get('MkblCarat', '').strip()  
        # SawLossCarat = request.form.get('SawLossCarat', '').strip()
        # CrapsCarat = request.form.get('CrapsCarat', '').strip()
        # EngGraphCts = request.form.get('EngGraphCts', '').strip()
        EngCts = request.form.get('EngCts', '').strip()
        EngShp = request.form.get('EngShp', '').strip()
        EngQua = request.form.get('EngQua', '').strip()
        EngCol = request.form.get('EngCol', '').strip()
        EngCut = request.form.get('EngCut', '').strip()
        EngPol = request.form.get('EngPol', '').strip()
        EngSym = request.form.get('EngSym', '').strip()
        EngFlo = request.form.get('EngFlo', '').strip()
        EngNts = request.form.get('EngNts', '').strip()
        EngMikly = request.form.get('EngMikly', '').strip()
        EngLab = request.form.get('EngLab', '').strip()
        EngAmt = request.form.get('EngAmt', '').strip()
        
        if EngLab == "nan":  # or if not EngLab:
            EngLab = np.nan
        else:
            EngLab = EngLab  # remains the same

        # Check if any required field is empty
        if not all([Tag, 
                    #ICarat, MkblCarat, SawLossCarat, CrapsCarat, EngGraphCts, 
                    EngCts, EngShp, EngQua, EngCol, 
                    EngCut, EngPol, EngSym, EngFlo, EngNts, EngMikly, 
                    EngLab, EngAmt]):
            flash("Please fill all the fields", "error")
            return redirect(url_for('home'))
        
        # Transform categorical features using loaded label encoders
        Tag = loaded_label_encoder['Tag'].transform([Tag])[0]
        EngShp = loaded_label_encoder['EngShp'].transform([EngShp])[0] 
        EngQua = loaded_label_encoder['EngQua'].transform([EngQua])[0]
        EngCol = loaded_label_encoder['EngCol'].transform([EngCol])[0]
        EngCut = loaded_label_encoder['EngCut'].transform([EngCut])[0]
        EngPol = loaded_label_encoder['EngPol'].transform([EngPol])[0]
        EngSym = loaded_label_encoder['EngSym'].transform([EngSym])[0]
        EngFlo = loaded_label_encoder['EngFlo'].transform([EngFlo])[0]
        EngNts = loaded_label_encoder['EngNts'].transform([EngNts])[0]
        EngMikly = loaded_label_encoder['EngMikly'].transform([EngMikly])[0]
        EngLab = loaded_label_encoder['EngLab'].transform([EngLab])[0]

        # Convert numeric fields to float
        # ICarat = float(ICarat)
        # MkblCarat = float(MkblCarat)
        # SawLossCarat = float(SawLossCarat)
        # CrapsCarat = float(CrapsCarat)
        # EngGraphCts = float(EngGraphCts)
        EngCts = float(EngCts)
        EngAmt = float(EngAmt)
        
        # Prepare input data for prediction
        input_features = [[Tag, 
                           #ICarat, MkblCarat, SawLossCarat, CrapsCarat, EngGraphCts,
                           EngCts, EngShp, EngQua, EngCol, EngCut, 
                           EngPol, EngSym, EngFlo, EngNts, EngMikly, EngLab, EngAmt]]
        
        # Make predictions
        gia_price = gia_model.predict(input_features)[0]
        grade_price = grade_model.predict(input_features)[0]
        bygrade_price = bygrade_model.predict(input_features)[0]
        makable_price = makable_model.predict(input_features)[0]
        
        gia_diff = EngAmt - gia_price
        grade_diff = EngAmt - grade_price 
        bygrade_diff = EngAmt - bygrade_price
        makable_diff = EngAmt - makable_price

    except Exception as e:
        flash(f"Error in calculating the price: {e}", "error")
        return redirect(url_for('home'))

    return render_template('results.html', gia_price=gia_price, grade_price=grade_price, 
                           bygrade_price=bygrade_price, makable_price=makable_price,
                           gia_diff=gia_diff,grade_diff=grade_diff,bygrade_diff=bygrade_diff,
                           makable_diff=makable_diff)

if __name__ == "__main__":
    app.run(debug=True, use_reloader=False)