FPBoost: a gradient boosting model for survival analysis that builds hazard functions as a combination of fully parametric hazards.
-
Updated
Oct 30, 2024 - Jupyter Notebook
FPBoost: a gradient boosting model for survival analysis that builds hazard functions as a combination of fully parametric hazards.
Competing-risks survival re-examination of the Mayo PBC cirrhosis benchmark (UCI Cirrhosis dataset): a leakage-free, reproducible pipeline benchmarking seven models against the Mayo clinical score.
An interactive Streamlit dashboard for analyzing equipment failure patterns and predicting maintenance needs. Features include data visualization, root cause analysis using Apriori algorithm, and survival modeling with Random Survival Forests to estimate time-to-failure and optimize spare parts management.
Random Survival Forest model for wildfire structure-proximity prediction · WiDS Global Datathon 2026 · Public score: 0.96373
Out of sample comparison of survival models (Cox PH, RSF, Gradient Boosting) under varying censoring rates.
Breast cancer survival analysis using Kaplan-Meier curves, Cox proportional hazards models, and Random Survival Forests.
This project explains "prompt engineering," a key technique for guiding AI models to desired outputs in tools like chatbots and text summarizers. It highlights the importance of clear instructions and techniques like CoT Prompting for effective communication with large language models. The project also introduces the Langchain library✨.
Two-stage pLTV framework for mobile gaming: Cox PH + Gradient Boosting survival models predict player churn risk; LightGBM + Random Forest predict 180-day revenue. 88 tests, 27 visualisations, stakeholder-ready docs.
PyTorch refactor of the TCAP (Transfer-learning based Cox Proportional Hazards Network) using scikit-survival.
Survival analysis for loan default timing using Cox PH and Random Survival Forest, served through FastAPI with a React dashboard
Reproducible pipeline for SUPPORT2 survival analysis with three-layer leakage audit, LODGO transportability benchmark, and bootstrap-validated baselines. PLOS ONE 2026.
Interpretable machine learning framework for predicting mid-term mortality and major periprocedural complications after transcatheter aortic valve implantation (TAVI) using pre-procedural clinical, echocardiographic, CT, and cusp-specific aortic valve calcium topography data. Includes reproducible survival and classification pipelines following TRI
Add a description, image, and links to the scikit-survival topic page so that developers can more easily learn about it.
To associate your repository with the scikit-survival topic, visit your repo's landing page and select "manage topics."