How I Deployed a Machine Learning Model for the First Time
Introduction
For as long as I’ve started with machine learning, Jupyter Notebooks have been my most loyal sidekick. From data preprocessing to model training, fine-tuning, and testing, Jupyter Notebooks have been there at every step to support me. However, I always knew that there is an entire world beyond these digital pages — a world of deployment and application.
Taking the leap from training a model to actually deploying it might seem intimidating. However, it’s a critical step that transforms a data science project from a theoretical experiment into a practical, real-world application. And I knew I had to take that extra step!
In this article, we will embark on my journey of building a classification model for a Kaggle competition. We start from a typical EDA and pipeline building until reaching new-unexplored territory — at least for me — bringing my machine learning model to life, enabling it to interact and offer insights to users globally.
Let’s brace ourselves as we step outside the comfort of our Jupyter Notebooks, because we’re about to go on a deployment journey. Grab your coding cap, fasten your seatbelt, and let’s get ready for a thrilling ride into the world of machine learning deployment!
Playground Series Episode 5 Season 3: Ordinal Regression with a Tabular Wine Quality Dataset
Our journey starts with the fifth episode of Kaggle’s Playground Series’ third season. This series, promoted by Kaggle, presents a variety of machine learning challenges, inviting users to boost their skills in data analysis, feature engineering, data cleansing, and machine learning pipeline construction.
0 Comments