Handwritten Character Recognizer

Try the app here : https://chetan-ml-app.herokuapp.com/

In this project, I will demonstratesteps to train and deploy a Machine Learning model with a web interface.

Tools required:

• A server for inference: Cloud instances, or simply your powerful laptop.
• Data: EMNIST Balanced an extended version of MNIST with 131,600 characters, 47 balanced classes.
• PyTorch: To train the deep learning model.
• Docker: Make our life easier to create a container for our application.
• Flask: For API and user interface.

Workflow:

The steps are as follows:

• Prepare the docker image using Dockerfile.
• Get data and train the machine learning model on Colab.
• Show a simple inference example.
• Build the API and the user-interface

Now we need to think about how to “ship” our model. This what we call packaging. It is simply to put the inference steps together and have a simple API to use it later.

If you would like to deploy your model on a cloud provider, I would suggest using free heroku servers.(limitations apply). Lately they started supporting docker images to be deployted as applications.

Folow these steps:

• heroku container:login
• heroku create
• heroku container:push web
• heroku container:release web
• heroku open