gdverse 1.6

• Fix stratification collision in interaction detection (#101).

• Loading gdverse no longer initializes the RNG state (#99).

• Update package startup message and citation guidance (#97).

• Report significance of interaction detection results in geographical detector (#95).

• Improve robustness of non-centrality parameter estimation for q-statistic confidence intervals (#93).

gdverse 1.5.1

• Wrap python-dependent examples in tryCatch to avoid check failures (#82).

gdverse 1.5

• Refactor cpd_disc() for improved parallel stability and reticulate compatibility (#77).

• Update python dependency configuration and usage (#74).

• Eliminate warnings about the size aesthetic for lines in plot S3 methods (#71).

gdverse 1.3-3

• Add pkg citation information (#61).

gdverse 1.3-2

• Experimentally support computing confidence intervals for the q statistic in factor_detector (#51).

• Support adjusting the font size of q-value labels in the factor detection result plot (#49).

• Rename gd_bestunidisc() as gd_optunidisc() (#39).

• Specify that the cores parameter in all functions of the gdverse package currently supports only positive integer inputs (#36).

• Update the documentation on Python dependency configuration in the rgdrid vignette (#30).

• Discuss the consistency with the results of existing geographical detector R packages (#28).

• Update the calculation method of pseudo p-values for the SPADE model (#26).

gdverse 1.3-1

• Fix bugs of gd,opgd,gozh,srsgd plot s3 methods (#23).

gdverse 1.3

• Redundant functions consolidated and migrated to sdsfun for maintenance (#22).

• Reorganize the parallel computing section in gdverse (#21).

• Align the RGD model with the original algorithm presented in paper (#20).

• Now gdverse requires sdsfun with a minimum version of 0.4.3 (#19).

• Transfer isp-related concepts and functions to the cisp package (#18).

• Modify the default font settings in the gdverse S3 plotting methods (#17).

gdverse 1.2

• Set the default number of discretizations in gdverse to range from 3 to 8 (#15).

• Optimize the Python integration setup in gdverse (#14).

• Now opgd() returns optimal discretization parameters (#13).

• Force data to tibble format in gdverse GDMs model function (#12).

• Align the RID model and algorithm with the original framework presented in paper (#9).

• Beautify the narrative and other writing details in the vignettes, without making any changes at the user level.

gdverse 1.1-1

• Clear the WORDLIST to ensure the source code remains clean and organized.

• Migrate the source code from ausgis/gdverse to stscl/gdverse on GitHub.

gdverse 1.1.0

• The general variable discretization in gdverse now utilizes sdsfun::discretize_vector() (#6).

• Algorithm functions are migrated to sdsfun (#8).

gdverse 1.0-3

• Update the RGD Model API Settings (#2).

• Fix bug caused by changes in default parameters of opgd in sesu_opgd (#4).

• The parameter overlaymethod in rid and idsa has been renamed to overlay.

• Add readr as a dependence of type Suggests.

• Recompile vignettes due to internal function changes.

gdverse 1.0-2

• When the discvar input for the opgd, rgd, rid, spade functions is NULL, it is assumed that all independent variables in the formula need to be discretized.

• Updating the S3 method for plotting various factor detectors to better conform to academic publication requirements.

gdverse 1.0-1

• Unify all vignettes filename to lowercase.

• Support for using the sf object as input in all models.

gdverse 1.0.0

• First stable release.