Flask/Pytorch/Docker starter app

This project is aimed to help machine learning developers to quickly build and deploy a Flask web app that take advantage of their machine learning ready PyTorch model. The documentation explains how to get up and running with either virtualenv or Docker.

This website is deployed in Heroku: https://flaskpytorch.herokuapp.com/

By default, this app uses MobileNetV2 image classifier that was pre-trained on the ImageNet dataset. This can be easily changed with any custom deep learning model.

Getting Started (using Python virtualenv)

You need to have Python installed in your computer.

1. Install virtualenv:

   pip install virtualenv
   

2. Create a Python virtual environment:

   virtualenv venv
   

3. Activate virtual environment:

   1. Windows:

   cd venv\Scripts
   activate
   cd ..\..
   

   2. Lunix / Mac:

   source venv/bin/activate
   

4. Install libraries:

   pip install -r requirements.txt
   

Run the code

• Run the app:

  flask run
  

• Run on a specific port:

  flask run -p <port>
  

Getting Started (using Docker)

1. Create a Docker image

   docker build -t pytorchflask .
   

   This will create an image with the name pytorchflask. You can replace that with a custom name for your app.

2. Run the docker image

   docker run -d -p 127.0.0.1:5000:80 pytorchflask
   

   This will run the app on port 5000. You can replace that with which ever port that is more suitable.

Deploying to Heroku

• Create Heroku app

  heroku create 
  git push heroku master
  

OR

• Add to existing Heroku app

  heroku git:remote -a <your-app-name>
  git push heroku master
  

Changing the model

1. Go to models.py
2. Follow the structure of the class MobileNetto create a custom model class
3. Use your class in app.py

Built With

• Pytorch - The Machine Learning framework used
• Flask - The web server library

Authors

• Imad Eddine Toubal - Initial work - imadtoubal

License

This project is licensed under the MIT License - see the LICENSE file for details

Happy coding!