scikit-survival-analysis
Time-to-event modeling with scikit-survival: Cox PH (elastic net), Random Survival Forests, Boosting, SVMs for censored data. C-index, Brier, time-dependent AUC; Kaplan-Meier, Nelson-Aalen, competing risks. Pipeline/GridSearchCV compatible. Use statsmodels for frequentist, pymc for Bayesian, lifelines for parametric.
npx skills add BioTender-max/awesome-bio-agent-skills --skill scikit-survival-analysis --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The skill directs an agent to perform time-to-event (survival) analysis using scikit-survival. It covers Cox proportional hazards models (including elastic-net penalization), ensemble survival methods (Random Survival Forest, Gradient Boosting, Extra Trees), survival SVMs (linear and kernel), non-parametric estimators (Kaplan-Meier, Nelson-Aalen), and competing risks. It emphasizes evaluation with censoring-aware metrics (C-index, Brier, time-dependent AUC), estimation of non-parametric curves, and CIFs for competing risks. It also notes compatibility with Pipelines and GridSearchCV, and references related tools for parametric or Bayesian approaches (lifelines, statsmodels, pymc).
How it works
- Preparation: Create structured survival arrays (event/time) and preprocess features (encode categoricals, standardize). Load data via provided utilities and ensure survival outcomes are in Surv format.
- Cox Proportional Hazards: Use CoxPHSurvivalAnalysis for standard Cox models; CoxnetSurvivalAnalysis for elastic-net penalized Cox with a path of alphas; IPCRidge provides accelerated failure time modeling.
- Ensemble Methods: Train RandomSurvivalForest, GradientBoostingSurvivalAnalysis, ComponentwiseGradientBoostingSurvivalAnalysis, or ExtraSurvivalTrees; obtain risk scores or survival functions, and derive survival curves.
- Survival SVMs: Train FastSurvivalSVM (linear) or FastKernelSurvivalSVM (kernelized) with standardized features; predict risk scores or labels; optional HingeLoss variant.
- Non-Parametric Estimation: Compute Kaplan-Meier and Nelson-Aalen estimates and plot; handle stratification by groups.
- Evaluation Metrics: Calculate concordance indices (Harrell and Uno), time-dependent AUC via cumulative_dynamic_auc, and Integrated Brier Score; compare model predictions against censored data.
- Competing Risks: Estimate Cumulative Incidence Functions with cumulative_incidence_competing_risks; fit cause-specific Cox models for individual events.
- Workflows: Provides example pipelines for standard survival pipelines and model comparison, including grid search over hyperparameters and evaluation on test data.
When to use it
- When modeling time-to-event outcomes with right-censoring in clinical trials or reliability studies.
- When you need Cox models (plain or elastic net), ensemble survival methods, or survival SVMs.
- When you require censoring-aware evaluation metrics and non-parametric survival curves.
- When exploring competing risks and cause-specific hazards.
What it can touch
- Libraries and modules referenced: scikit-survival, scikit-learn, numpy, pandas, matplotlib.
- Models and classes: CoxPHSurvivalAnalysis, CoxnetSurvivalAnalysis, IPCRidge, RandomSurvivalForest, GradientBoostingSurvivalAnalysis, ComponentwiseGradientBoostingSurvivalAnalysis, ExtraSurvivalTrees, FastSurvivalSVM, FastKernelSurvivalSVM, HingeLossSurvivalSVM, ClinicalKernelTransform, kaplan_meier_estimator, nelson_aalen_estimator, cumulative_incidence_competing_risks, concordance_index_ipcw, cumulative_dynamic_auc, integrated_brier_score, Surv, load_breast_cancer, load_gbsg2, load_arff.
Caveats
- The skill description assumes right-censoring handling; left- or interval-censoring are noted as better supported by other tools (e.g., lifelines).
- It documents usage with Python versions and dependencies, but actual environment compatibility depends on the execution context.
- Licenses: the skill metadata lists GPL-3.0; confirm license compatibility with your project.
# scikit-survival -- Survival Analysis ## Overview scikit-survival is a Python library for time-to-event analysis built on scikit-learn. It handles right-censored data (observations where the event has not yet occurred) using Cox models, ensemble methods, survival SVMs, and non-parametric estimators. All models follow the scikit-learn `fit/predict` API and integrate with Pipelines, cross-validation, and GridSearchCV. ## When to Use - Modeling time-to-event outcomes with right-censored data (clinical trials, reliability) - Fitting Cox proportional hazards models (standard or elastic net penalized) - Building ensemble survival models (Random Survival Forest, Gradient Boosting) - Training survival SVMs for margin-based learning on medium-sized datasets - Evaluating survival predictions with censoring-aware metrics (C-index, Brier score, AUC) - Estimating non-parametric survival curves (Kaplan-Meier, Nelson-Aalen) - Analyzing competing risks with cumulative incidence functions - High-dimensional survival data with automatic feature selection (CoxNet L1/L2) - For **simpler parametric models** (Weibull, log-normal AFT) or statistical tests (log-rank), use `lifelines` - For **deep learnin
- Overview
- When to Use
- Prerequisites
- Quick Start
- Core API
- Module 1: Data Preparation
- Module 2: Cox Proportional Hazards
- Module 3: Ensemble Methods
- Module 4: Survival SVMs
- Module 5: Non-Parametric Estimation
- Module 6: Evaluation Metrics
- Module 7: Competing Risks
- Key Concepts
- Model Selection Guide
pip install scikit-survival scikit-learn pandas numpy matplotlib
What does the scikit-survival-analysis skill do?
Time-to-event modeling with scikit-survival: Cox PH (elastic net), Random Survival Forests, Boosting, SVMs for censored data. C-index, Brier, time-dependent AUC; Kaplan-Meier, Nelson-Aalen, competing risks. Pipeline/GridSearchCV compatible. Use statsmodels for frequentist, pymc for Bayesian, lifelines for parametric.
How do I install it?
Run `npx skills add BioTender-max/awesome-bio-agent-skills --skill scikit-survival-analysis --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From BioTender-max/awesome-bio-agent-skills, a repository with 135 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.