Deep Learning for Tabular Data using PyTorch Tabular
In spite of showing unreasonable effectiveness in modalities like Text and Image, Deep Learning has always lagged Gradient Boosting in tabular data—both in popularity and performance. But recently there have been newer models created specifically for tabular data, which is pushing the performance bar. But popularity is still a challenge because there is no easy, ready-to-use library like Sci-Kit Learn for deep learning.
PyTorch Tabular is a new deep learning library which makes working with Deep Learning and tabular data easy and fast. It is a library built on top of PyTorch and PyTorch Lightning and works on pandas dataframes directly.
Many SOTA models like NODE and TabNet are already integrated and implemented in the library with a unified API. PyTorch Tabular is designed to be easily extensible for researchers, simple for practitioners, and robust in industrial deployments.
The library is available at https://github.com/manujosephv/pytorch_tabular
- Tabular Data and Deep Learning - 2 mins
- The journey through SOTA - 5 mins
- Design Principles of PyTorch Tabular - 5 mins
- Unified API for SOTA Tabular Models - 5 mins
- Usage Examples - 3 mins
- Advanced Usage and Tricks - 8- mins
- How to install and contribute? - 2 mins
This is geared towards absolute beginners to Deep Learning as well as those who are an an intermediate to advanced level. the focus is on tabular data. which is the most popular type of data out here(csv, sql tables, etc.)
Basic Machine Learning
Manu Joseph is a Lead Data Scientist at Thoucentric, where he leads the Deep Learning Practice. He has worked in the industry for more than 8 years in a wide variety of roles ranging from Software Engineer to Supply Chain consultant and finally as Data Scientist. His primary research interests lie in applying Deep Learning for Time Series Forecasting and Natural Language Processing. He is also passionate about learning and spreading knowledge and writes in his own blog at https://deep-and-shallow.com/.
Lead Developer: PyTorch Tabular : https://github.com/manujosephv/pytorch_tabular
NLTK Contributor: https://github.com/nltk/nltk/tree/develop/nltk/lm