linafaik08/survival_analysis

How linafaik08/survival_analysis works

This is an experimentation project for survival analysis, with two use cases: customer subscription churn and hospital readmission. The documented workflow runs preprocessing, exploration, survival-model training, and evaluation notebooks in order; the project also includes reusable splitting, ML and deep-learning training, and scoring code. The README supports the notebook workflow and saved model-score outputs. Notebook internals and their specific calls into src were not sampled, so those implementation links are omitted.

Main entry points: 03_modeling_survival_customer_subscription.ipynb Customer modeling · 13_modeling_survival_hospitalisation.ipynb Hospital modeling · 01_data_preprocessing_customer_subscription.ipynb Customer preprocessing · 11_data_preprocessing_hospitalisation.ipynb Hospital preprocessing

Components 17

Use-case data

Subscription data datasets/customer_subscription
Hospital data datasets/hospitalisation

Experiment workflow

Customer preprocessing notebooks/01_data_preprocessing_customer_subscription.ipynb
Customer exploration notebooks/02_data_exploration_customer_subscription.ipynb
Customer modeling notebooks/03_modeling_survival_customer_subscription.ipynb
Customer evaluation notebooks/04_evaluation_customer_subscription.ipynb
Hospital preprocessing notebooks/11_data_preprocessing_hospitalisation.ipynb
Hospital exploration notebooks/12_data_exploration_hospitalisation.ipynb
Hospital modeling notebooks/13_modeling_survival_hospitalisation.ipynb
Hospital evaluation notebooks/14_evaluation_hospitalisation.ipynb

Training and evaluation code

Data splitting src/train.py
Survival ML src/train_survival_ml.py
Deep survival src/train_survival_deep.py
Score analysis src/evaluate.py

Saved results

Saved models outputs/models
Model scores outputs

Other components

Experimenter

Connections 14

Experimenter → Customer preprocessing: runs first README.md
Subscription data → Customer preprocessing: prepares README.md
Customer preprocessing → Customer exploration: then explores README.md
Customer exploration → Customer modeling: then trains README.md
Customer modeling → Customer evaluation: then evaluates README.md
Customer modeling → Saved models: saves models README.md
Customer evaluation → Model scores: records scores README.md
Experimenter → Hospital preprocessing: runs first README.md
Hospital data → Hospital preprocessing: prepares README.md
Hospital preprocessing → Hospital exploration: then explores README.md
Hospital exploration → Hospital modeling: then trains README.md
Hospital modeling → Hospital evaluation: then evaluates README.md
Hospital modeling → Saved models: saves models README.md
Hospital evaluation → Model scores: records scores README.md

Drawn by GitDiagram on November 9, 2026 from the repository's file tree, README and source files. Read as Markdown ·