---

CoxKAN Documentation

CoxKAN is a Python package for performing survival analysis using Kolmogorov-Arnold Networks.

Contents:

• CoxKAN Introductory Demo
• API
• Datasets Subpackage

Indices and tables

• Index

• Module Index

• Search Page

Installation

To install CoxKAN, use pip:

pip install coxkan

Usage

Here is a simple example of how to use CoxKAN:

from coxkan import CoxKAN
from coxkan.datasets import metabric
from sklearn.model_selection import train_test_split

df = metabric.load()
df_train, df_test = train_test_split(df, test_size=0.2)
dataset_name, duration_col, event_col, covariates = dataset.metadata()

ckan = CoxKAN(width=[len(covariates),2,1], grid=5, k=3)

log = ckan.train(
    df_train,
    df_test,
    duration_col='duration',
    event_col='event',
    opt='Adam',
    lr=0.01,
    steps=100)

For more information, visit the CoxKAN GitHub repository.

Contributing

Contributions welcome! Please submit a pull request or open an issue on GitHub.