Madrigalpp / Torch-version-for-TCGA-data-DeepSurv-

realization DeepSurv and Coxnnet in TCGA mi-RNA analysis

Deep-Learning-for-survival-analysis Pytorch

realization DeepSurv and Coxnnet in TCGA mi-RNA analysis
This project's paper includes a section dedicated to its deep learning analysis, specifically focusing on the implementation of the XGBENC deep learning analysis： XGBENC

Overview

This project uses deep learning methods to implement prognostic analysis of TCGA-miRNA. We propose a scalable code framework covering grid search, 5-fold cross-validation methods. This project analyzes two veteran survival analysis algorithms. DeepSurv and Coxnnet and give model evaluation.

Requirment

Pytorch>=0.4.0
CPU or GPU

<pip install requirements.txt>

Data-available

you should download from TCGA and convert to s-g_data100_data.csv format. TCGA

How to use ？

you can start in you CMD via
<python main.py>

network.py contains all the Network settings and Partial Likelihood loss function
ini_file.py contain all the hyper parameters
utils.py contain the c-index calculation and other settings
you can run DL_survival_main to run the model.

Reference

czifan/DeepSurv.pytorch ([e-mail])
if you have any problems, please contact Wankang Zhai ([e-mail])