The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
Initial public release.
metahunt() chains denoised functional SPA basis
hunting, constrained simplex projection, and Dirichlet weight modelling
in a single call. Method dispatch for predict(),
summary(), and plot() on the returned
"metahunt" object.split_conformal() and cross_conformal()
return distribution-free prediction intervals around the target function
(pointwise on the grid or, with a wrapper, around a scalar
summary).conformal_from_fit() adds intervals to an already-fit
pipeline using a held-out calibration set.coverage(), summary(), and
plot() methods for the "metahunt_conformal"
class.reconstruction_error_curve() (unsupervised elbow) and
cv_error_curve() (supervised CV) for picking
K.select_denoising_params() cross-validates the
(N, Delta) knobs of dfspa().dfspa() denoised functional Successive Projection
Algorithm (Algorithm 1 of the paper).project_to_simplex() constrained simplex projection of
each study’s function onto the recovered bases (quadratic program via
quadprog).fit_weight_model() and
predict.metahunt_weight_model() for Dirichlet regression of
simplex weights on study-level covariates, with
coef.metahunt_weight_model() for inspecting
coefficients.predict_target() and apply_wrapper() for
composing predictions and scalar summaries by hand.build_grid() constructs a shared evaluation grid from
any reference patient-level dataset.f_hat_from_models() evaluates a list of fitted models
on the shared grid with class-aware dispatch for ranger,
grf (causal_forest,
regression_forest), and a default branch that covers
lm/glm/randomForest. Custom S4
classes can supply their own predict_fn.minmax_regret() implements the covariate-free
worst-case-regret aggregator of Zhang, Huang, and Imai (2024, arXiv:2412.11136).metahunt-intro, data-prep,
grid-weights, wrapper-scalar, plus
get-started.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.