Changelog

mapac 0.31

• aa_gtable() Fix bug that prevents the conversion of zero values to “-” in the last two rows.

mapac 0.30

• aa_card() sets negative UA standard errors to zero if the absolute value is less than the specified precision (se_precision = 1e-9) to avoid NA’s.

mapac 0.29

• New aa_recode() to recode confusion matrices to a new classification scheme. aa_confusion_matrix_recode() is deprecated and removed.
• New aa_lut() to create a lookup table from labeled vectors.

mapac 0.28

• aa_flextable() and aa_gtable() add capitalization option.
• aa_gtable() turns off cell colors with cell_colors = FALSE.
• aa_word() exports and appends confusion matrices to new or existing Word document.
• aa_card() uses argument h or named vector w to match stratum weights with the right classes.
• aa_card() and aa_stratified() add new element “adjustment” to output. It is set to TRUE if weights were supplied.

mapac 0.27

• aa_card() changed input arguments. Previously, it was possible to submit reference and map vectors with cbind(), which converts factors to characters not the way we want. This change breaks backwards compatibility.
• aa_stratified() adds default values for h and N_h in. This assumes a proportional sampling design.

mapac 0.26

• New aa_flextable() and aa_flextable_classic() can save to Word documents.
• New aa_gtable() but keep aa_confusion_matrix_gtable() for backwards compatibility.
• Moved package pages to gitlab CI.
• Housekeeping.

mapac 0.25

• aa_confusion_matrix_gtable() fixed coloring.

mapac 0.24

• aa_card() adds F-1 score and associated standard errors.

mapac 0.23

• Newaa_sample_distance() for sub-sampling with a minimum distance.

mapac 0.21

• rma() removed cancor and prediction accuracy scores form because they may be misinterpreted if rma is applied to predicted vs observed values.

mapac 0.20

• New rma() function for reduced major axis regression.

mapac 0.12

• aa_confusion_matrix_gtable() update matches gt package update.

mapac 0.11

• aa_confusion_matrix_gtable() fixed error after gt package update

mapac 0.10

• New aa_sample_size() to estimate sample size.
• aa_stratified() adds a warning if one or more strata have a sample size of 1.

mapac 0.09

• aa_stratified() adds finite population correction factor.
• aa_confusion_matrix_gtable() adapted confusion matrix colors.
• Add unit test.

mapac 0.08

• aa_stratified() adds F-1 score and associated standard errors.
• Add vignettes
• Change to pkgdown documentation

mapac 0.07

• aa_card() fixes bug that used Card formula even if olofsson==T. The bug only affected standard error etimates of the reference class proportions and the standard error of the overall accuracy. The differences are negligible.