Changes in version 1.1.0 (2026-06-10)                  

Major rewrite of the package for a configurable, dual-backend
neural-network imputer.

New features

  - Architecture is now user-configurable via deepimp_arch() (number and
    size of hidden layers, dropout, activation, batch normalisation,
    optimiser, learning rate), replacing the previously hard-coded
    network.
  - Two backends: a native torch backend (default; no Python required)
    and an optional keras3 backend, selected with backend = "torch" /
    "keras".
  - impNNetCoDa() reimplemented on the shared imputation engine for
    rounded zeros in compositional data, with correction = "truncate"
    (default, Templ 2021), "expectation", or "none".
  - Results are returned as a "deepimp" object; read the completed data
    with getImputed() (or as.data.frame()), and inspect with print(),
    summary(), and plot().
  - Reproducible runs via a seed argument. Reproducibility is exact when
    dropout = 0; with dropout > 0 the backend's training-time randomness
    is not fully pinned by the seed in the current torch build (repeated
    runs are close but may not be bit-identical).
  - A vignette, vignette("deepImp"), and a package overview page
    (?deepImp).

Bug fixes

  - impNNetCoDa(initialize = "kNNa") no longer aborts at high
    rounded-zero rates. The compositional-kNN initialiser
    (robCompositions::impKNNa()) can fail with 'probs' not in [0, 1]
    when a large fraction of every part falls below the detection limit;
    the initialiser now falls back to knn (with a warning) instead of
    stopping the imputation.

Deprecations and breaking changes

  - impNNet() was redesigned. Legacy arguments dense, small,
    normalize_x, and normalize_y are soft-deprecated (a warning maps
    them onto the new API). The arguments model_numeric, model_binary,
    model_nominal, and loss were removed; configure the network through
    arch and backend instead.

Notes

  - With m > 1, impNNet() / impNNetCoDa() return several completed data
    sets. These are stochastic replicate completions, not statistically
    valid multiple imputation.