Scooter is an experimental, scalable RESTful API for serving machine learning (ML) models. It's framework agnostic in the sense that you can use whatever ML you prefer. All that is required is you implement a process that handles your predictions.

Scooter comes with example code for serving Keras models.

Scooter is inspired by https://www.pyimagesearch.com/2018/02/05/deep-learning-production-keras-redis-flask-apache/.

Installing Scooter does not automatically install Tensorflow, Keras or similar. Again, it's totally up to you to choose what frameworks (if any) you want to use for prediction.

Before you can use Scooter, you must install Redis. This can be done as follows: