somepago/saint
A PyTorch implementation of SAINT, a deep learning model for tabular data using row attention and contrastive pre-training.
Not currently ranked — collecting fresh signals.
star history
This repository provides the official implementation of the SAINT paper, introducing improved neural network architectures for tabular data. The model employs row-level attention mechanisms and contrastive pre-training strategies to enhance performance on tabular classification and regression tasks. Built in PyTorch, it supports various attention variants including column-only, row-only, and combined approaches, and can leverage unlabeled data for unsupervised pretraining to improve performance with limited labeled samples.
Frequently asked
- What is somepago/saint?
- A PyTorch implementation of SAINT, a deep learning model for tabular data using row attention and contrastive pre-training.
- Is saint open source?
- Yes — somepago/saint is open source, released under the Apache-2.0 license.
- What language is saint written in?
- somepago/saint is primarily written in Python.
- How popular is saint?
- somepago/saint has 465 stars on GitHub.
- Where can I find saint?
- somepago/saint is on GitHub at https://github.com/somepago/saint.