Best Practices in Feature Engineering for Tabular Data With GPU Acceleration

Feature engineering is an important component in (tabular) machine learning solutions, which can be easily integrated into existing models and boost model accuracy. The tabular data structure limits the models' capabilities to learn the relationships between features, so adding handcrafted features can significantly improve model performance. We'll teach best practices for feature engineering techniques specific to tabular data building off our teams' collective experience competing in data science competitions such as Kaggle and RecSys. Learn how to create features from categorical, numerical, and time-series data, and accelerate your data frame operations on GPU. We'll use RAPIDS, an open-source software that accelerates the whole data science pipeline from data preprocess/engineering to machine learning on GPU.
Prerequisite(s):

Python programming using libraries, such as pandas and NumPy.
Basic understanding of machine learning, decision trees and feed forward neural networks.