Perform survival analysis on censored data with Cox models and ensemble methods.
npx skills add https://github.com/ovachiever/droid-tings --skill scikit-survival
Please help me install this Agent Skill. Skill: scikit-survival Source: https://github.com/ovachiever/droid-tings/tree/main/skills/scikit-survival Command: npx skills add https://github.com/ovachiever/droid-tings --skill scikit-survival
Comprehensive survival analysis tooling in Python, including Cox models, ensemble methods, survival SVMs, and evaluation metrics.
Load survival data, fit a model (e.g., CoxPHSurvivalAnalysis), and evaluate predictive performance.
High-intent search queries and answers about installing and using this skill.
Survival analysis models time-to-event outcomes when some observations are censored. Use scikit-survival to load data as Surv objects, fit models like Cox proportional hazards or ensemble methods (Random Survival Forest, Gradient Boosting), and evaluate with concordance index or time-dependent AUC to predict risk scores and survival functions.
Cox proportional hazards model hazard ratios parametrically; Random Survival Forest and Gradient Boosting capture non-linear interactions through ensemble splits; Survival SVMs optimize margin-based ranking. Choose Cox for interpretability, ensembles for predictive performance on complex data, and SVMs for distance-based risk stratification.
Concordance index (C-index) measures ranking accuracy of risk predictions. Time-dependent AUC captures discrimination at specific time points. Brier score quantifies prediction error for survival probabilities. Use all three to assess calibration, discrimination, and overall predictive accuracy across your time-to-event model.
Yes, scikit-survival supports competing risks workflows where multiple event types can occur. Model each risk separately or jointly using appropriate Surv object specifications, enabling prediction of event-specific cumulative incidence and risk stratification in complex clinical or observational settings.
Preprocess with imputation for missing values, encoding for categorical features, and scaling for numeric predictors. Handle time and event columns correctly in Surv objects, then fit and validate models. Proper preprocessing ensures model convergence and reliable risk score generation.
kjgarzaModel time-to-event data with censoring using scikit-survival in Python.
virajpConfigure pnpm workspace and npm settings for secure JavaScript monorepos.
sanand0Guide classical machine learning tasks with the scikit-learn Python library.
xjtulycFit Kaplan-Meier, Cox, and AFT models to right-censored time-to-event data.
