For companies that depend on machine learning, it is very important to deploy their models in a scalable fashion. The models should work in the same manner that they were working in when the model was created. Deploying fastText models, both supervised and unsupervised, can be done in a variety of ways. The choice of methodology would depend on your individual needs.
In this chapter, we will focus on how to deploy fastText models in web and mobile scenarios. The topics are the following:
- Deploying to the web
- Flask
- FastText functions
- Flask endpoints
- Deploying to smaller devices
- Prerequisites – Completing the Google tutorial
- App considerations
- Adding the fastText model
- FastText in Java
- Adding library dependencies to Android
- Using library dependencies in Android
- Finally, the app
