Machine learning models could have tremendous value only when delivered to the end-users. The end-user could be recommender systems in the real-estate platform that suggests properties to renters or investors — Zillow, for instance.
However, machine learning projects can only be successful when a model is deployed, and its predictions are being served.
I was surprised that the machine learning deployment is unusually discussed online — this particular skill you need to learn in the practice workflow.
I tried to google this particular topic, but I found many blog posts about setting Flask APIs for machine learning models. However, none of these tutorials go into detail ahead, developing the only endpoint.
So I decided to blog about this topic in a comprehensive tutorial series on how to deploy ML models into production. I would start by conducting primary statistical analyses using the Jupyter notebook, then building a customized machine learning framework — package. After that, I will develop an endpoint API using FLASK APP. Finally, I would use the CI/CD pipeline to deploy the machine learning model to the Paas Platform.
Some of the technologies we would use are Docker, Gemfury, Flask-API, CircleCI, Kaggle API, and Sklearn. So excited, let’s begin the journey.
_Disclaimer, The tutorial is more focused on how things work, not a code line-by-line tutorial. At any point, you can use the _Github repo_ commit history to reference your code and, of course, ask if you need help._
This is a two parts tutorial, this is part one which includes build and publish a machine learning python package. Part two includes building a Flask API end points and deploying to Heroku.
Please note this is an intermediate tutorial, you need to meet certain requirements to be able to catch up. However, I tried to be clear as possible as I can by adding comments and attach resources for further learning and reading.
In Part one , I will walk you through the steps required to develop your own machine learning framework to automate building steps from fetching the dataset to publishing to the cloud so you can download and use.
#deployment-automation #data-science #machine-learning #ml-with-sklearn #framework
How to deploy your first Machine learning models. Develop customized ML pipelines from describing the business problem to deployment. This extensive guide includes Docker and packaging configurations.