Metadata-Version: 2.1
Name: discotime
Version: 0.1.0
Summary: Discrete-time competing risk analysis with neural networks
Home-page: https://github.com/peterchristofferholm/discotime
Author: Peter Holm
Author-email: "Peter Holm" <petchdk@gmail.com>
License: MIT
Classifier: License :: OSI Approved :: MIT License
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE

[![Documentation Status](https://readthedocs.org/projects/discotime/badge/?version=latest)](https://discotime.readthedocs.io/en/latest/?badge=latest)

# Discotime: Discrete-time Competing Risks Models with PyTorch Lightning

[**Docs**](https://discotime.readthedocs.io/en/stable/)
| [**Installation**](https://discotime.readthedocs.io/en/stable/installation.html)
| [**Tutorial**](https://discotime.readthedocs.io/en/stable/tutorial/index.html)
| [**Examples**](https://github.com/discotime/examples)

*Discotime* is a python package for discrete-time survival analysis 
with neural networks that have been designed to handle competing risk models.
The package relies on PyTorch Lightning to provide an easy-to-use interface,
that still can be customized to your heart's content. 
The packages contains an implementation of discrete time-to-event models
for neural networks (using PyTorch), different evaluation metrics,
and a couple of different competing risk datasets. 

**The package is currently under development,**
**Breaking changes are to be expected.**

