Changes in version 0.3.0                        

  - The defaults of boots() and batts() are now the settings of the
    paper's simulations. boots(): gradient boosting (use_gradient =
    TRUE), at most 1000 trees (num_trees_max) selected by 5-fold CV
    (K_CV = 5) and per-tree subsampling of one half (subsample_fraction
    = 0.5); depth 4, learning rate 0.01, 32 bins and margin 0.1 are
    unchanged. batts(): 200 trees (num_trees), 2000 burn-in iterations
    (size_burnin) and 1000 saved draws (size_backfitting, previously
    required); the other defaults are unchanged.
  - Calls that relied on the previous defaults now give different
    results and use the random number generator for CV folds and
    subsamples. Supply the previous values explicitly to keep earlier
    behaviour. Estimation code is unchanged; fits with the same explicit
    settings are identical to 0.2.2.
  - The paper's simulations used 0.2.2 (tag v0.2.2) with explicit
    settings.
  - The README example and example.R use the new defaults, with a
    shorter MCMC run (500 burn-in iterations and 500 draws) for a quick
    demonstration.

                        Changes in version 0.2.2                        

  - Add subsample_fraction to boots() (default 1, unchanged behavior and
    RNG use). A value below one builds each tree from a random subsample
    of ceiling(subsample_fraction * m) training observations, drawn as
    in ada with bag.frac: one observation per group, the rest uniformly
    without replacement regardless of group. Splits, gradients and leaf
    values use the subsample; all weights are updated, and the
    forward-stagewise normalization and CV losses use the whole training
    set and held-out folds, respectively.
  - The internal run_adaboost() entry point takes the per-tree subsample
    indices as an additional argument; batts() does not subsample.

                        Changes in version 0.2.1                        

  - Reject invalid burn-in and retained-draw counts before fitting
    (U22).
  - Reject move probabilities outside the documented positive,
    length-three, unit-sum contract in R and C++; the sum tolerance
    is 1e-12 (U23).
  - These guards preserve valid-input calculations and RNG consumption.
    Invalid inputs now fail before fitting. Inverse-Gaussian limitation
    U16 is unchanged.

                        Changes in version 0.2.0                        

  - Require a positive integer thin and validate prediction matrix
    dimensions and finite values before passing data to C++ (U14, U17).
  - Breaking API change: remove update_lambda, lambda_prior_parameters,
    and alpha_cutpoint. Leaf scale stays fixed at num_trees * lambda_0;
    cut spacing/proposals use the previous default. Remove these named
    arguments from existing calls. Nondefault cut grids and scale
    updating are no longer supported. The C++ interface is regenerated
    with the reduced argument list.
  - Document inverse-Gaussian cancellation under extreme parameters as
    known limitation U16; its numerical formula is unchanged by author
    decision.

                        Changes in version 0.1.0                        

Official distribution: https://github.com/nawaya040/BATTS.

Fixes integrated from development

This release integrates commits 6ea6722 through 909ea357 after the
previous default-branch commit 6f625ba. They correct forward-stagewise
normalization and prediction constants, posterior-forest storage and
guards, CV group handling, gradient CV scaling, split loss and
split-boundary rules, PRUNE indices, and empty-leaf sampling and GROW
moves. They add input guards and restrict Bayesian initialization to
unsplit trees. They also reduce unnecessary copies, history allocation
and internal-node index storage.

These existing changes can alter estimates and RNG streams. A run using
0.1.0 must record its own version and settings; numerical identity with
the previous public version or every reported paper result is not
asserted.

Additional release fixes

  - Rename the prediction-only Node type to PredictionNode, removing its
    one-definition-rule collision with the fitting Node type.
  - Require C++17 for compatibility with recent RcppArmadillo.
  - Replace placeholder author/maintainer metadata, add API help and
    regression tests, and document the official release installation.

The new type-name fix preserves mathematical operations and RNG calls.
Build-level equivalence and limitations are documented in
inst/VALIDATION.md. No model, prior, canonical experiment setting or
saved result is changed by this release preparation.