Upgini: A Game Changer Python Library for Feature Enrichment in Machine Learning
Uplift the accuracy of your machine learning models in a few simple steps.
Data is the fuel of all machine learning models. The more data we have, the more accurate and robust models we can create. Even if we managed to build a state-of-the-art machine learning algorithm, the models will show what is in the data. Thus, it is not a surprise that the number one suggestion for improving a model is always gathering more data if possible.
Gathering more data is not an easy task and we might need to spend hours to come up with new informative features. This is where Upgini comes into play.
What is Upgini?
Upgini is a feature search and enrichment library in Python. It searches through thousands of public and community data sources and then creates features that enrich the capabilities of our models. These new features improve the prediction power of the models.
Upgini covers a variety of data sources both public and shared by the data science community such as historical weather, international holidays, consumer confidence index, world house prices, public social media profile data, and so on. Although the amount of data being searched is very large, the results come up very quickly. We can have a new…
