cobsurv - Cobra Ensemble for Conditional Survival

cobsurv is a project stands for Cobra Ensemble for Conditional Survival which serve the python package cobsurv , cobsurv is a python package for survival analysis based on the proximity measure. Currently it host only one method COBRA Survival but in future we will add different methods based on the proximity measure. We will also provide natively build Random Forest and Newarest Neighbour Survival, currently the package uses the [scikit-survival](sebp/scikit-survival) and [npsurvival](georgehc/npsurvival) for Random Forest and Nearest Neighbour Survival respectively.

The objective of the package to establish State of the Art proximity based cobsurv to predict the individual survival function and time dependent effect of covariates.

Install

The easiest way to install cobsurv is to use Python’s package manager pip

pip install cobsurv

A jupyter notebook to start with COBRA Survival

This page contain, only one example of COBRA Survival right now but later we will add more examples of different methods.

API Reference

The reference guide contains a detailed description of the cobsurv API. It describes which classes and functions are available and what their parameters are.

Contributing

If you come across a typo in the documentation or have ideas for adding new functionalities to cobsurv, we welcome your contributions! The contributing guidelines will assist you in the process of setting up a development environment and submitting your changes to the cobsurv team