0

I have created a Flask prediction app (within Google Colab) and when I am trying to run it post adding all the dependencies within the colab environment I am getting the url but when I click on it it show site cannot be reached.

I have the Procfile, the pickled model and the requirements text file but for some reason its not working. Also, I tried deploying this app using Heroku and it met the same fate where I got the app error.

For more context please visit my github repo.

Any help or guidance will be highly appreciated.

from flask import Flask, url_for, redirect, render_template, jsonify
from pycaret.classification import*
import pandas as pd
import numpy as np
import pickle

app = Flask(__name__)
model = load_model('Final RF Model 23JUL2021')
cols = ['AHT','NTT','Sentiment','Complaints','Repeats']

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

@app.route('/predict',methods=['POST'])
def predict():
    int_features = [x for x in request.form.values()]
    final = np.array(int_features)
    data_unseen = pd.DataFrame([finak], columns = cols)
    prediction = predict_model(model, data=data_unseen, round=0)
    prediction = int(prediction.Label[0])
    return render_template('home.html',pred='Predicted Maturiy Level is{}'.format(prediction))

@app.route('/predict_api',methods=['POST'])
def predict_api():
    data = request.get_json(force=True)
    data_unseen = pd.DataFrame([data])
    prediction = predict_model(model, data=data_unseen)
    output = prediction.Label[0]
    return jsonify(output)

    if __name__ == '__main__':
    app.run(debug=True)
Sunil Kappal
  • 1
  • 2

1 Answers1

0

You cannot run flask app same as in your machine. You need to use flask-ngrok.

!pip install flask-ngrok

from flask_ngrok import run_with_ngrok
[...]
app = Flask(__name__)
run_with_ngrok(app)
[...]
app.run()

You can't use debug=True parameter in ngrok.

charchit
  • 1,492
  • 2
  • 6
  • 17