Changes in version 2025.5.13 (2025-05-27) o correct first/last subtrain.borders (was defined using first/last validation data, now using all/train data). Changes in version 2025.5.6 o add new container: STL priority_queue (heap), almost no difference with multiset. Changes in version 2025.4.29 o update Rcpp code in interface.cpp to avoid access warnings: use .begin() instead of &vec[0] to get pointer to first element of vector. Changes in version 2023.10.24 o laplace test 1:8 yields asymmetric split. Changes in version 2023.8.31 (2023-09-06) o update un-exported function arg docs to avoid CRAN NOTE. Changes in version 2022.7.21 (2022-08-07) o init max_zero_var=0 to avoid valgrind msg. Changes in version 2022.7.19 (2022-07-20) o New log-linear cum_median function. o New NEWS format to please CRAN. o rm deprecated binary_function in PiecewiseFunction.h. Changes in version 2022.7.13 o binseg: constant factor speed improvements, mostly by going back to storing params in class members rather than unordered_map. Changes in version 2022.4.14 o Function re-naming. o New QP solver for tree viz. o Changes in version 2022.4.13 o New R function get_best_optimal implements dynamic programming for finding the tree with smallest number of splits for a given set of input sizes (N.data, min.segment.length, o C++ depth_first method computes fast best case number of splits, optimal when computing full path, heuristic when segments < data. Changes in version 2022.4.6 o complexity funs handle min segment length. o operator< now correctly breaks ties: previously used segment size (end-start), now use max distance from start and end, to encourage equal splits and best case time complexity. Changes in version 2022.4.4 o new l1 and laplace loss functions. Changes in version 2022.3.30 o comparisons vignette. o max_zero_var computed based on the max estimated variance of all single data points, which should be zero, but are sometimes small non-zero values (for example 1e-15 or 1e-13) due to numerical issues. mean_zero_var used in meanvar_norm loss function to determine if cost is finite. Segments with an infinite cost best split are not stored in the container for later splitting. Changes in version 2022.3.29 o meanvar_norm distribution: generalize C++ code to more than one segment-specific parameter. Changes in version 2022.3.24 o container.str can be list (slow) or multimap (fast). Changes in version 2022.3.22 o warning and suggestion to use weights for runs of identical data. o R code binseg(min.segment.length=3) etc uses min_segment_length parameter in C++ code. Changes in version 2022.3.11 o new binseg function with distribution=poisson or mean_norm, weight.vec. Changes in version 2022.1.24 (2022-01-26) o remove random_set_vec test.rev example which failed on M1. Changes in version 2021.11.3 (2022-01-20) o binseg_normal_cv does model selection via most frequent number of segments with minimum validation error (over several random splits). Changes in version 2021.11.2 o break ties in Segment operator< by size (split larger segments first). o binseg_normal gains args is.validation.vec, position.vec in order to support efficient cross-validation. it now returns list with new component subtrain.borders (predicted changepoint positions). o get_splits* functions for comparing empirical to best/worst case. Changes in version 2021.1.6 o More comments in binseg_normal.cpp to help potential GSOC students. Changes in version 2020.10.7 o Comment binseg_normal.cpp to explain optimal_cost computation. o Use C++ multiset with operator< instead of multimap/vector. o Use cumsum C++ vector for constant time mean/cost computation for any split. o Store cost of segments before/after split, pass the cost values to maybe_add to avoid having to recompute them. o Computation works for only one data point. Changes in version 2020.9.15 o remove unused C++ errors. o test coef method. Changes in version 2020.9.3 (2020-09-14) o Bugfix for negative means, docs. o predict/plot methods, copy code from example. Changes in version 2019.9.20 o Initial implementation.