neuralsbi 0.3.2

README is now generated from README.Rmd, so the usage example runs at render time and its output cannot drift from the code. The example is a plain linear regression: the posterior recovers the ground-truth coefficients, which the reader can check against ordinary least squares.

neuralsbi 0.3.1

CRAN resubmission fixes. Routed the summary() and as.data.frame() methods into the single summaries help topic (they had drifted into separate .Rd files with duplicated \alias entries, which also produced duplicate HTML anchors). Wrapped theta_{<d} in \eqn{} in the made_masks docs so the Rd no longer drops braces. Dropped the bare “NPE” acronym from the DESCRIPTION to avoid the spurious misspelling note.

Defaults now match Python sbi, so a workflow reads the same in both packages and results can be cross-checked. Changes to npe() defaults: the density estimator is now "maf" (was "mdn"); MDN mixture components default to 10 (was 5); NSF spline bins default to 10 (was 8); the training batch size is 200 (was 100). max_epochs is raised to 2000 as a guard cap that early stopping (patience = 20) normally reaches first, mirroring sbi’s effectively-unbounded epoch budget. lr, validation_fraction, patience, clip_grad_norm, n_transforms, and hidden already matched. Pass any of these explicitly to recover the previous behavior.
First CRAN submission. Dropped the development .9000 version suffix, removed the redundant Author/Maintainer fields (now derived from Authors@R), and tidied the package title.
Embedding networks (roadmap v0.4). embedding_mlp() builds a learned summary network that maps raw observations to a low-dimensional feature vector; pass it to npe(..., embedding_net = ) and the MDN, MAF, and NSF estimators condition on the features instead of the raw data, training the embedding jointly. The estimators still take raw x at the de_* boundary (dim_x is unchanged), so sampling and log_prob route through the embedding automatically. Ignored, with a warning, by linear_gaussian.

Vignettes now show real output. They are precomputed: each vignette’s evaluated source lives in vignettes/<name>.Rmd.orig, and vignettes/precompute.R bakes it into a static vignettes/<name>.Rmd (results, printed values, and figures inlined). CI and pkgdown re-render that static Markdown with no torch at build time, so the expensive neural training runs once, locally, instead of on every build. Re-run Rscript vignettes/precompute.R after editing any .Rmd.orig.
Two-moons calibration study (inst/benchmarks/two_moons_calibration.R): SBC, expected coverage, and TARP for a two-moons NSF fit, with figures written to docs/figures/ (roadmap milestone M2).

Package website built with pkgdown, deployed from CI to https://pedroliman.github.io/neuralsbi/.
Four vignettes that build on each other: getting started, choosing a density estimator, checking the posterior, and the SIR case study (which now also demonstrates npe_sequential()). Removed a truncated duplicate of the SIR vignette.
README rewritten to the standard terse form; authorship recorded in DESCRIPTION (Pedro Nascimento de Lima, with ORCID).

New npe_sequential(): multi-round NPE targeting a single observation via truncated-prior proposals (TSNPE, Deistler et al. 2022). Each round truncates the prior to the highest-probability region of the current posterior and retrains on all accumulated simulations; the standard NPE loss stays valid, so no importance correction is needed. Returns an nsbi_snpe fit that works with posterior(), sample(), and the diagnostics, but is only valid at the targeted x_obs. Verified against the analytic linear-Gaussian posterior.

New tarp() diagnostic and plot_tarp() (Lemos et al. 2023): a joint expected-coverage test using random reference points, complementing the per-parameter sbc() ranks. Detects posteriors with calibrated marginals but wrong correlation structure.
New plot_posterior_predictive(): marginal predictive histograms with the observation marked; returns the observation’s predictive quantiles.
Leakage correction is now under test: with a bounded prior, the renormalized log_prob() integrates to one over the support and returns -Inf outside it (test-posterior-normalization.R).
Fixed CI. R CMD check failed on three counts: the npe() example required libtorch (it now uses the torch-free linear_gaussian estimator and runs unconditionally), the hand-maintained npe.Rd/fit_mdn.Rd usage sections had drifted behind the code (missing n_restarts, clip_grad_norm, n_transforms, and the "maf"/"nsf" options), and CLAUDE.md was not in .Rbuildignore. The test-torch job also failed because torch 0.17 refuses a TORCH_HOME that does not exist; the workflow now creates it first.

Shared training engine for all neural estimators (train_conditional_de()): best-of-n restarts, learning-rate decay on plateau, gradient clipping, per-epoch loss history.
Masked Autoregressive Flow (density_estimator = "maf") and Neural Spline Flow ("nsf", autoregressive rational-quadratic splines) join the MDN and the closed-form linear_gaussian baseline.
Benchmark tasks (task_gaussian_linear(), task_two_moons(), task_slcp(), task_sir()) shared between tests and the inst/benchmarks/ head-to-head benchmark harness.
summary() methods, as.data.frame() tidy accessor, plot_coverage().
SIR applied case-study vignette.
CI: R CMD check plus a test-torch job with cached libtorch.

First pilot release: priors, single-round amortized npe(), linear_gaussian and MDN estimators, posterior sampling with leakage correction, SBC, expected coverage, C2ST, posterior-predictive checks, pairplot(), plot_sbc().