bartMachine - Bayesian Additive Regression Trees
Fits Bayesian additive regression trees (BART), a Bayesian nonparametric machine-learning method, for regression and classification. Provides posterior inference, credible and prediction intervals, variable selection, interaction analysis, partial-dependence plots, cross-validation, and optional graphics processing unit (GPU) acceleration through TornadoVM.
Last updated
bayesian additive regression treesbartbayesian machine learningbayesian nonparametricsmachine learningensemble learningdecision treesregressionclassificationposterior inferencemcmcvariable selectionuncertainty quantificationinterpretable machine learningpartial dependencecross-validationgpu accelerationbayesian-additive-regression-treesbayesian-machine-learningbayesian-nonparametricsbayesian-statisticsdecision-treesensemble-learninggpu-accelerationinterpretable-machine-learningmachine-learningpartial-dependenceposterior-inferenceuncertainty-quantificationvariable-selectionopenjdk
11.07 score 67 stars 3 dependents 327 scripts 3.9k downloads
EDI - Experimental Design and Inference
Implements a comprehensive suite of experimental designs, both fixed (e.g., block, stratified, matched-pair, cluster, factorial, and mixed-integer-programming-based optimal designs) and sequential (including matching-on-the-fly designs, biased coin designs, and covariate-adaptive urn designs that assign treatment one subject at a time while maintaining covariate balance), for continuous, incidence, count, proportion, survival, and ordinal response types. For each design and response type combination, provides the corresponding inference procedures, including exact, asymptotic, distribution-free, and resampling-based (bootstrap, jackknife, and randomization) methods, so that estimation and testing are always matched to how the data were generated. An 'InferenceSuite' facility runs all applicable inference procedures for a given design and response type at once and reports a single Cauchy-combined p-value summarizing their evidence. A built-in simulation framework supports power analysis and operating-characteristic studies across designs, response types, and inference procedures, with optional parallelization via 'mirai'. Missing covariate data is handled automatically via built-in imputation. Core numerical routines are implemented in C++ via 'Rcpp' for speed on large designs and simulation studies. Machine-specific tuning for optimization is included.
Last updated
causal-inferenceclinical-trialscppexperimental-designpythonrandomization-inferencesequential-designstatisticssurvival-analysisopenblascppopenmp
7.39 score 1 stars 1.2k scripts 248 downloadsbartMachineJARs - bartMachine JARs
These are bartMachine's Java dependency libraries. Note: this package has no functionality of its own and should not be installed as a standalone package without bartMachine.
Last updated
bartbayesian-additive-regression-treesbayesian-machine-learningbayesian-nonparametricsbayesian-statisticsclassificationcross-validationdecision-treesensemble-learninggpu-accelerationinterpretable-machine-learningmachine-learningmcmcpartial-dependenceposterior-inferenceregressionuncertainty-quantificationvariable-selectionopenjdk
6.57 score 67 stars 4 dependents 8 scripts 3.1k downloads
ICEbox - Individual Conditional Expectation Plot Toolbox
Implements Individual Conditional Expectation (ICE) plots for visualizing models fitted by supervised learning algorithms. ICE plots refine Friedman's partial dependence plot by graphing the fitted relationship between the predicted response and a feature for individual observations. They reveal variation in fitted values across the feature's range, helping identify where and to what extent interactions and other heterogeneous effects may be present.
Last updated
individual conditional expectationice plotsderivative ice plotsd-ice plotspartial dependence plotsinterpretable machine learningmodel interpretabilityexplainable aifeature effect visualizationinteraction detectionheterogeneous effectsblack-box modelssupervised learningstatistical learningblack-box-modelsdata-visualizationderivative-iceexplainable-aifeature-effectsheterogeneous-effectsice-plotsindividual-conditional-expectationinteraction-detectioninterpretable-machine-learningmachine-learningmodel-interpretabilitypartial-dependencestatistical-learningcppopenmp
5.54 score 35 stars 60 scripts 1.6k downloads
PTE - Personalized Treatment Evaluator
Provides out-of-sample inference for evaluating whether a pre-specified personalized treatment recommendation model improves outcomes relative to business-as-usual treatment assignment. Supports continuous, binary, and survival endpoints from two-arm randomized trials, cross-validated performance estimates, confidence intervals and hypothesis tests based on the bootstrap, and data-driven selection of the m-out-of-n bootstrap size. Default model-fitting kernels are implemented in C++ through 'Rcpp' for computational efficiency.
Last updated
personalized medicineprecision medicinepersonalized treatment recommendationsindividualized treatment rulestreatment regimesheterogeneous treatment effectstreatment policy evaluationpolicy value estimationcomparative effectiveness researchcausal inferencerandomized controlled trialsclinical trialsstatistical inferencebootstrap inferencem-out-of-n bootstrapcross-validationsurvival analysisr packagebootstrapcausal-inferenceclinical-trialscomparative-effectivenessheterogeneous-treatment-effectsindividualized-treatment-rulespersonalized-medicinepersonalized-treatmentpolicy-evaluationprecision-medicinerandomized-controlled-trialsstatistical-inferencesurvival-analysistreatment-policy-evaluationtreatment-regimescppopenmp
4.74 score 1 stars 41 scripts 335 downloads
GreedyExperimentalDesign - Greedy Experimental Design Construction
Computes experimental designs for two-arm experiments with covariates using multiple methods, including: (0) complete randomization and randomization with forced-balance; (1) greedy optimization of a balance objective function via pairwise switching; (2) numerical optimization via 'gurobi'; (3) rerandomization; (4) Karp's method for one covariate; (5) exhaustive enumeration for small sample sizes; (6) binary pair matching using 'nbpMatching'; (7) binary pair matching plus method (1) to further optimize balance; (8) binary pair matching plus method (3) to further optimize balance; (9) Hadamard designs; and (10) simultaneous multiple kernels. For the greedy, rerandomization, and related methods, three objective functions are supported: Mahalanobis distance, standardized sums of absolute differences, and kernel distances via the 'kernlab' library. This package is the result of a stream of research that can be found in Krieger, A. M., Azriel, D. A., and Kapelner, A. (2019). "Nearly Random Designs with Greatly Improved Balance." Biometrika 106(3), 695-701 <doi:10.1093/biomet/asz026>. Krieger, A. M., Azriel, D. A., and Kapelner, A. (2023). "Better experimental design by hybridizing binary matching with imbalance optimization." Canadian Journal of Statistics, 51(1), 275-292 <doi:10.1002/cjs.11685>.
Last updated
experimental designrandomizationcovariate balancererandomizationmatchingoptimizationpower analysisstatistical powercppopenjdk
4.31 score 27 scripts 477 downloadsbclogit - Conditional Logistic Regression
Performs inference for Bayesian conditional logistic regression with informative priors built from the concordant pair data. We include many options to build the priors. And we include many options during the inference step for estimation, testing and confidence set creation. For details, see Kapelner and Tennenbaum (2026) "Improved Conditional Logistic Regression using Information in Concordant Pairs with Software" <doi:10.48550/arXiv.2602.08212>.
Last updated
bayesian-statisticsconditional-logistic-regressionmatched-case-control-studiesmatched-pairsconcordant-pairsinformative-priorsepidemiologybiostatisticsbinary-outcomesstanonetbbcpp
3.93 score 1 stars 6 scripts 258 downloads
YARF - Yet Another Random Forests Package
A customizable, asynchronous, and parallelized random forest implementation. Supports user-defined splitting and aggregation behavior, missingness incorporated in attributes, and random-forest-based imputation.
Last updated
random-forestdecision-treesensemble-learningmachine-learningstatistical-learningsupervised-learningclassificationregressionvariable-importanceinteraction-detectionproximity-analysisout-of-bagmissing-datamissing-data-imputationcustom-split-rulesasynchronous-computingparallel-computingjavajavascriptopenjdk
3.56 score 1 stars 36 scripts
fastLogisticRegressionWrap - Fast Logistic Regression Wrapper
Provides very fast logistic regression with coefficient inferences plus other useful methods such as a forward stepwise model generator (see the benchmarks by visiting the github page at the URL below). The inputs are flexible enough to accomodate GPU computations. The coefficient estimation employs the fastLR() method in the 'RcppNumerical' package by Yixuan Qiu et al. This package allows their work to be more useful to a wider community that consumes inference.
Last updated
logistic-regressiongeneralized-linear-modelsbinary-classificationstatistical-inferencemaximum-likelihoodmodel-selectionstepwise-selectionaicpredictionhigh-performance-computinggpu-computingrcppcpp
3.48 score 1 stars 2 dependents 3 scripts 431 downloadsYARFJARs - YARF JARs
These are YARF's Java dependency libraries. Note: this package has no functionality of its own and should not be installed as a standalone package without YARF.
Last updated
asynchronous-computingclassificationcustom-split-rulesdecision-treesensemble-learninginteraction-detectionjavajavascriptmachine-learningmissing-datamissing-data-imputationout-of-bagparallel-computingproximity-analysisrandom-forestregressionstatistical-learningsupervised-learningvariable-importanceopenjdk
2.48 score 1 stars 1 dependents
optDesignSlopeInt - Optimal Designs for Estimating the Slope Divided by the Intercept
Aids practitioners to optimally design experiments that measure the slope divided by the intercept and provides confidence intervals for the ratio.
Last updated
optimal experimental designdesign of experimentslinear regressionratio estimationslope-to-intercept ratioconfidence intervalscalibrationhenrys law constant
1.70 score 2 scripts 221 downloads