Skip to content

Get Started

This website contains the documentation for the package bornrule available on PyPI. The package implements the classifier proposed in the paper Text Classification with Born's Rule. All code is available at the GitHub repository.

Installation

Install via pip with:

pip install bornrule

Usage

The package implements three versions of the classifier. The classification algorithm is compatible with the scikit-learn ecosystem. The neural version is compatible with pytorch. The SQL version supports in-database classification.

Scikit-Learn

from bornrule import BornClassifier
  • Use it as any other sklearn classifier
  • Supports both dense and sparse input and GPU-accelerated computing via cupy

PyTorch

from bornrule.torch import Born
  • Use it as any other torch layer
  • Supports real and complex-valued inputs. Outputs probabilities in the range [0, 1]

SQL

from bornrule.sql import BornClassifierSQL
  • Equivalent to the class BornClassifier but for in-database classification
  • Supports inputs represented as json {feature: value, ...}

Cite as

Emanuele Guidotti and Alfio Ferrara. Text Classification with Born's Rule. In Advances in Neural Information Processing Systems, volume 35, pages 30990–31001, 2022.

A BibTeX entry for LaTeX users is:

@inproceedings{guidotti2022text,
 title = {Text Classification with Born's Rule},
 author = {Guidotti, Emanuele and Ferrara, Alfio},
 booktitle = {Advances in Neural Information Processing Systems},
 pages = {30990--31001}, 
 volume = {35},
 year = {2022}
}