BMEmapping 2.0.0
BMEmapping 2.0.0 (2026-07)
Major Features
- Quantile-Based BME (QBME) Integration: Added full
pipeline support for the parameter-free QBME interpolation engine via
q_bme_predict(), q_bme_predict_ci(),
q_bme_cv(), and q_prob_zk().
- Native S3 Plotting Architecture: Extended the base
S3
plot() and summary() methods to accept
prediction, cross-validation, and density collection objects directly,
enabling continuous surface and uncertainty field mapping.
Minor Improvements & Bug
Fixes
- Explicitly namespaced base statistical functions
(
stats::approx(), stats::density()) to
eliminate namespace collision warnings during package check.
- Replaced direct unquoted column selection in
ggplot2 internal aesthetics with the
rlang::.data pronoun wrapper to prevent unbound global
variable flags.
- Pruned obsolete package dependencies (
utils) from the
Imports profile in the DESCRIPTION namespace
to satisfy tidy deployment standards.
- Fixed an issue in R Markdown compilation where raw LaTeX
\url{} expressions failed to generate clean hyperlinks
across HTML and docx output engines.
BMEmapping 1.2.2
- Fixed minor namespace exports and structural input validations for
the Classical BME pipeline.
BMEmapping 1.2.0
- Optimization of the integration windows during continuous spatial
probability support profiling.
BMEmapping 1.0.0
- Stable baseline release of the classical BME interpolation
suite.
BMEmapping 0.3.0
Major Features
- Extended the internal integration
zk_range window
bounds by 20% to prevent density truncations and provide more robust
non-Gaussian localized support estimates.
- Re-engineered covariance matrix construction functions to
structurally preserve spatial continuity signatures under heavy
nugget-effect penalties.
Minor Improvements & Bug
Fixes
- Flattened multi-dimensional distance arrays to eliminate intensive
Kronecker-product tensor operations, generating massive execution
speedups during global coordinate matrix evaluations.
- Tightened input sanitization checks to reject inconsistent
dimensions or incomplete spatial objects prior to entropy
maximization.
- Refined standard logging profiles to provide clear error messages
for degenerate data or ill-conditioned covariance structures.
BMEmapping 0.2.0
- Refined input validation pipelines and expanded warning triggers for
singular matrices.
BMEmapping 0.1.0
Major Features
- Initial core release providing full computational framework support
for Bayesian Maximum Entropy spatial interpolation across hard
observations and soft-interval data domains.
Minor Improvements & Bug
Fixes
- Initial CRAN-ready codebase structures.
- Implemented clean custom printing and tabular text summary methods
for BME model class assignments.
- Restructured package boundaries to streamline the package vignette
footprint and drop bloated optional dependencies.
- Added Dr. Yan Sun as package author.
- Hardened internal numerical stability tolerances during matrix
decompressions and localized density profiling.