Changes in version 1.7.5 (2025-12-06) o New vignette "Adaptive fitting" o `disclapmix_adaptive()`: - Allowing to take `label` and generators for `init_y` (`init_y_generator`) and `init_v` (`init_v_generator`) - refer to the "Adaptive fitting" vignette. - Allowing `disclapmix_adaptive()` to take multiple `criteria`'s - Not necessary to supply `_marginal` suffix for `criteria` o Add `cluster` argument to `simulate()` to force simulating from a particular cluster o `disclapmix()` for `glm_method` `glm.fit`: `covmat` calculated more efficiently o Parameters for `disclapmix_robust()` controlling retries with random subsets o Include check for if data contains `NA`'s o New `init_y_method`: `hclust` o `predict` can now accept NA values, in which case that locus is ignored (corresponding to marginalising over that locus). o Remove "fwsim" Suggest o Add `export_compact_fit` method for easier exporting fits for prediction (e.g., to JSON) o CRAN fixes Changes in version 1.7.4 (2022-06-29) o `disclapmix()` takes new agument: `init_v` (initial posterior probabilities to get EM started) o `disclapmix()` takes new agument: `ret_x` of whether to return data or not -- the default is not to (as opposed to earlier behaviour) o Added function `disclapmix_robust()` which is a wrapper around `disclapmix()` that tries to avoid errors. Can sometimes avoid errors with svd problems. o Added function `disclapmix_adaptive()` which is a wrapper around `disclapmix_robust()` that instead of fitting one model for a given number of clusters, fits models until the best model (lowest marginal BIC) is in the interior (with margin `M`) of all number of clusters tried. E.g., the best model has 3 clusters and the margin `M = 5`, then this function ensures that models with 1, 2, ..., 3+5 = 8 clusters are fitted. If e.g. then 7 is better than 3, then it continues such that also models with up to 7+5 = 12 clusters are fitted. Changes in version 1.7.3 (2019-03-12) o Adding suppressWarnings(RNGversion("3.5.0")) to tests as a result of https://bugs.r-project.org/bugzilla/show_bug.cgi?id=17494. Changes in version 1.7.2 (2019-01-09) o Maintenance updates o Remove Makevars files as not required by Rcpp >= 0.11. Changes in version 1.7.1 o Tag for the The Journal of Open Source Software paper: http://joss.theoj.org/papers/10.21105/joss.00748 Changes in version 1.7 (2018-04-22) o Moving to Github o Roxygen o testthat o Travis CI Changes in version 1.6.3 o Avoid two centers being equal (mainly a problem for very few loci) o Internal: PedigreeSums added with increased stack size and progress indicator Changes in version 1.6.2 (2015-09-09) o Minor technical changes for keeping CRAN checks happy Changes in version 1.6.1 (2015-06-12) o Plot functionality for disclapmix was broken for fits with only one subpopulation/cluster. Changes in version 1.6 (2015-01-16) o Plot functionality for disclapmix fit takes an argument for the distances between clusters (defaults to clusterdist(x)). This can be used if the distances have been precalculated. o Minor technical changes for keeping CRAN checks happy Changes in version 1.5 (2014-06-14) o Added AICc (Akaike Information Criterium with finite sample correction) o Added plot functionality for a disclapmix fit o Corrected number of model observations, leading to corrected BIC values (AIC not affected) o Error in full likelihood ratio corrected (not used for model selection) Changes in version 1.4 (2014-04-25) o Added separation of two persons mixtures (see example in ?rank_contributor_pairs) o Added AIC (Akaike Information Criterium) for a model disclapmixfit Changes in version 1.3 (2014-04-07) o Added simulate from fitted model o Added haplotype diversity calculation from a fitted model o Added parameter to control number of IRLS iterations (glm_control_maxit) o Added parameter to control IRLS convergence (glm_control_eps) o Changed default iterations to 100L, glm_control_maxit to 50L and glm_control_eps to 1e-6 o Fixed error with no IRLS output for verbose = 2L expect for glm.fit o Small changes in verbose output Changes in version 1.2 (2013-10-16) o Changed default glm_method to internal_coef due to speed considerations o Fixed issues when having one cluster, one locus or both Changes in version 1.1 (2013-10-11) o Code changes for OS X, Windows and Solaris compatibility Changes in version 1.0 (2013-10-10) o Totally rewritten such that the algorithm is more memory and CPU efficient and can be used to analyse larger datasets o Internal IRLS algorithm implemented, much faster than glm.fit for this kind of data. glm.fit is still available. See glm_method in ?disclapmix. o NOTE: Updated API Changes in version 0.4 o Fixed error when there is no variation in a subpopulation Changes in version 0.3 (2013-03-06) o Fixed error such that verbose = 1 outputs progress information o Number of model parameters corrected, which changes marginal and full BIC slightly, but only with a constant change meaning that for the same data, the optimal number of centers is the same Changes in version 0.2 (2012-04-16) o disclapmix: verbose more fine grained o FIXED: disclapmix: use.parallel = TRUE created problems under Windows o Corrected minor documentation errors Changes in version 0.1 (2012-04-13) o Initial release