main.py

# flask , scikit-learn, pandas, pickle-mixin
import pandas as pd
from flask import Flask, render_template, request
import pickle
import numpy as np

app = Flask(__name__)

data = pd.read_csv("Cleaned_data.csv")
pipe = pickle.load(open("RidgeModel.pkl", 'rb'))


@app.route('/')
def index():

    locations = sorted(data['location'].unique())
    return render_template('index.html', locations=locations)


@app.route('/predict', methods=['POST'])
def predict():
    location = request.form.get("location")
    bhk = request.form.get('bhk')
    bath = request.form.get('bath')
    sqft = request.form.get('totally sqft')

    print(location, bhk, bath, sqft)
    input = pd.DataFrame([[location, sqft, bath, bhk]], columns=['location', 'total_sqft', 'bath', 'bhk'])
    prediction = pipe.predict(input)[0]*1e5

    return str(prediction)


if __name__ == "__main__":
    app.run(debug=True, port=5000)