Changes in version 1.2-5 (2024-10-11)                  

  - Improved family parsing and code touch ups.

                 Changes in version 1.2-4 (2024-04-29)                  

  - Fixed opt_boost()

                 Changes in version 1.2-3 (2024-03-18)                  

  - Added the modified Cholesky family.

  - Some touch ups for gamlss.dist family objects.

                 Changes in version 1.2-2 (2023-10-29)                  

  - More support for deep distributional networks.

  - Small touch ups for ff based smooth terms.

                 Changes in version 1.2-1 (2023-07-04)                  

  - Support for distributions3/topmodels infrastructure.

                 Changes in version 1.2-0 (2023-06-13)                  

  - Removing maptools dependencies.

  - Fixes in JM code.

  - Improvements on estimating deep distributional neural networks using
    TensorFlow.

                 Changes in version 1.1-9 (2022-11-09)                  

  - Small fix in C code.

                 Changes in version 1.1-8 (2022-06-10)                  

  - Removing akima and geoR dependencies.

  - Fixed predict problems when using by variables.

                 Changes in version 1.1-7 (2022-04-08)                  

  - Full support for scaleable distributional regression models.

  - Bug fixing/C code.

                 Changes in version 1.1-6 (2021-11-24)                  

  - JSS paper published https://www.jstatsoft.org/article/view/v100i04.

                 Changes in version 1.1-5 (2021-09-21)                  

  - Bug fixing.

                 Changes in version 1.1-4 (2021-05-15)                  

  - Improved batchwise backfitting algorithm. Data is not copied
    anymore.

  - New setup for neural network model terms.

  - Bug fixing.

                 Changes in version 1.1-3 (2021-01-25)                  

  - New naming convention for optimizer and sampling functions.
    Optimizer are now indicated with prefix opt_, sampler functions with
    prefix sam_. Old optimizer and sampler functions are still supported
    for reverse compatibility.

  - New function engines(), which lists available optimizer and sampling
    functions for families.

  - New function CRPS() for computing the continuous rank probability
    score.

  - Bug fixing.

                 Changes in version 1.1-2 (2020-02-19)                  

  - Bug fixing.

  - Removed bit package depends.

  - More stable implementation of the batchwise backfitting optimizer.

                 Changes in version 1.1-1 (2019-11-13)                  

  - Bug fixing.

  - Support new features for generating weights for neural networks.

                 Changes in version 1.1-0 (2019-09-27)                  

  - New features.

  - Bug fixing.

  - New paper introducing bamlss at https://arxiv.org/abs/1909.11784

  - In addition, there is now a new website http://www.bamlss.org/ with
    technical examples, textbook examples, as well as examples from
    publications. The website will be expanded in the future and will
    serve as the main source for presenting new features of the bamlss
    package.

  - Adding a vignette to CRAN which links to the new website.

                 Changes in version 1.0-2 (2019-07-30)                  

  - Bug fixing.

  - First neural network implementations, the n() constructor.

  - Support for very large data sets using the ff and ffbase package.

  - Experimental batchwise backfitting algorithm bbfit().

                 Changes in version 1.0-1 (2018-10-12)                  

  - Bug fixing.

  - More model fitting engines: lasso(), stabsel()

                 Changes in version 0.1-2 (2017-04-14)                  

  - Fixed some issues with tx() for BayesX.

  - Added tx3() for BayesX.

  - Fixed environment saving problem when using light = TRUE in
    bamlss().

  - Dropped some parts of the jm.mode() return value.

  - Solved PROTECT() problems in the C code.

  - Many other small edits ...