The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
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[@]metanet.ch.
gg_cv() (cross-validation diagnostics),
gg_bootstrap() (confidence interval forest plot),
gg_weights() (weight sensitivity heatmap).villa_romana dataset: realistic 300-find Roman villa
with 4 phases, bioturbation, construction cuts, and residual
pottery.cv_sef(),
optimize_weights()).reorder_phases()
applied to each replicate before ARI computation, solving label
switching in confidence intervals.neighbourhood is set, energy is divided by actual neighbour
count instead of n-1, removing peripheral bias.feature_matrix() now
accepts external center/scale parameters and
preserves scaling attributes for proper cross-validation.adjusted_rand_index() and
confusion_matrix() for comparing estimated vs true
phases.us_summary_table() aggregates
diagnostics per stratigraphic unit;
phase_transition_matrix() reveals vertical phase ordering;
export_results() writes all results to CSV.harris_from_contexts()
auto-generates penalties from depth ordering; read_harris()
imports CSV edge lists; validate_phases_harris() checks
phase-stratigraphy consistency.gg_convergence() (EM trace),
gg_phase_profile() (depth vs phase),
gg_confusion() (confusion matrix heatmap with ARI).bootstrap_sef() for bootstrap confidence intervals on
PDI, entropy, energy, loglik, and ARI.cv_sef() for k-fold cross-validation of phase count
selection.optimize_weights() for data-driven SEI weight
estimation via grid search cross-validation.reorder_phases() relabels phases by mean depth (phase 1
= deepest/oldest) to solve label switching.sei_sparse() wrapper for sparse SEI computation on
large datasets (n > 1000).fit_sef() gains n_init parameter for
multiple random initialisations (default: 1). Best run by log-likelihood
is retained.fit_sef() default em_iter increased from
25 to 100 for better convergence.$converged flag in
sef_fit objects; warning issued on non-convergence.fit_sef() and compare_k() model selection
metrics.sei_matrix() gains max_dist parameter for
sparse computation (zeroes beyond threshold).sei_matrix() and ese() fully vectorized
(50-100x faster on large datasets).as_plotly() with
enriched archaeological tooltips (ID, context, phase, dating, class,
entropy, energy, intrusion probability).gg_* functions now embed tooltip data for seamless
plotly conversion.@examples,
@family, and @seealso for all exported
functions._pkgdown.yml for documentation website.cran-comments.md for CRAN submission.demo_easy,
demo_moderate, demo_compressed.summary.sef_fit() and
print.sef_fit() (NAMESPACE now uses S3method()
instead of export()).as_sf_phase() and as_sf_links() CRS
default (NA → NA_integer_) for compatibility
with sf >= 1.0.Authors@R in DESCRIPTION.LazyData field from DESCRIPTION.fit_sef().as_sf_links() and enhanced
as_sf_phase() with XY/XYZ support.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.