add_vec_mat_gpu         Broadcast-add a vector to a matrix, GPU-aware
calculateMu             Compute fitted means from a negative binomial
                        GLM fit
checkGPU                Is an accelerated (GPU) torch device in use?
diag_mat                Build a diagonal matrix from a vector,
                        GPU-aware
dnbinom_gpu             Negative binomial density (mean/size
                        parameterisation), GPU-aware
fastSizeFactors         Filter genes based on expression
filterGenes             Filter genes based on expression
fitNB                   Fit a per-gene negative binomial GLM
getBackendDevice        Resolve the active torch backend device
getBackendDtype         Resolve the active torch backend dtype
getGPUMemoryBudget      Determine the accelerator memory budget for
                        blocked fitting
gpuDtypeBytes           Bytes per element for the active backend dtype
hasBadValues            Does an object hold any NA/NaN/Inf?
HumanDLPFC              Human dorsolateral prefrontal cortex (DLPFC)
                        visium sample
invert_mat              Invert a symmetric positive-definite matrix
invert_mat_batched      Invert a batch of symmetric positive-definite
                        matrices
is_torch_tensor         Is an object a torch tensor?
isPolished              Does a SpaNormFit carry a converged per-gene
                        polish?
matmul_gpu              Tensor-aware matrix multiplication
mult_vec_mat_gpu        Broadcast-multiply a vector and a matrix,
                        GPU-aware
nbBlockDesign           A design with a per-group block, in compact
                        form
nbDevianceMoments       First two moments of the NB unit deviance, by
                        direct summation
nbUnitDeviance          NB unit deviance
plotCovariate           Diagnostic plot of predicted expression for a
                        covariate
plotSpatial             Plot spatial transcriptomic annotations per
                        spot
polishNB                Converge every gene of a shared negative
                        binomial fit to its own optimum
polishSpaNorm           Converge each gene of a SpaNorm fit to its own
                        optimum
qlDispersion            Per-gene quasi-likelihood dispersion
qlMomentTable           A shared table of NB deviance moments
resetGPUCache           Clear the cached GPU device/memory-budget state
rowSums_gpu             Row sums, GPU-aware
setGPUMemoryBudget      Set the accelerator memory budget for blocked
                        fitting
SpaNorm                 Spatially-dependent normalisation for spatial
                        transcriptomics data
SpaNormFit              An S4 class to store a SpaNorm model fit
SpaNormPCA              GLM-based (SpaNorm) PCA
SpaNormSVG              Model-based spatially variable gene (SVG)
                        calling
tcrossprod_gpu          Tensor-aware 'tcrossprod'
toGPUMatrix             Convert a matrix to a torch tensor on the
                        active backend
toGPUVector             Convert a vector to a torch tensor on the
                        active backend
topSVGs                 Export top SVG results to a data frame
toRMatrix               Convert a torch tensor (or R object) to a base
                        R matrix
tpsBasis                Thin-plate-style tensor spline basis over
                        spatial coordinates
