Deployment of the image classification DL model for techlabs Berlin project titled Beautify Berlin

Running the app locally

You can run the app locally running the following commands in the terminal

Clone the repo

git clone https://github.com/allakulov/techlabs_flask_app

Set up a Python virtual environment (if it doesn't exist)

sudo apt install python3.8-venv
sudo apt-get update python3.8-venv

Create a virtual environment

sudo python3 -m venv venv
source venv/bin/activate

Install requirements

Please run the following commands to install the requirements/dependencies

pip install -r requirements.txt

Launch application locally

python app.py

Deploy the app on Heroku

Create a Procfile in the main directory with the following line

web: gunicorn wsgi:app

Create a wsgi.py file with the following line

from main import app

if __name__ == “__main__”: 
    app.run()

Install Heroku CLI by following the instruction in this link

https://devcenter.heroku.com/articles/heroku-cli

Initialize a git repository, add and commit files

git init 
git add .
git commit -m "Initial commit"

Log in to Heroku CLI

heroku login

Create an app on Heroku

heroku create <choose-app-name>

Push files to Heroku remote

git push heroku master

Once your files are successfully pushed to Heroku, it will take a few minutes for the app to be built and launched.