Struct HistogramPlan 

pub struct HistogramPlan<T: Element> { /* private fields */ }

histogram plan.

1-D uniform-bin histogram via atomic-bin accumulation (PyTorch torch.histc). Bins cover [lo, hi] uniformly.

When to use: forward 1-D histogram. No BW (set-valued).

Dtypes: input {f32, f64}; output always i32 counts.

Shape limits: input flat [numel]; output [num_bins]. hi ≥ lo, num_bins ≥ 1.

Workspace: none. Launcher zeros output before the kernel runs.

Precision guarantee: non-deterministic — atomic-bin accumulation order varies across launches. Final counts are data-determined.

Implementations

impl<T: Element> HistogramPlan<T>

pub fn select( _stream: &Stream, desc: &HistogramDescriptor, _pref: PlanPreference, ) -> Result<Self>

Pick a kernel for desc.

pub fn can_implement(&self, args: &HistogramArgs<'_, T>) -> Result<()>

Validate args.

pub fn workspace_size(&self) -> usize

Workspace size in bytes.

pub fn sku(&self) -> KernelSku

Identity of the kernel this plan picked.

pub fn precision_guarantee(&self) -> PrecisionGuarantee

Numerical guarantees for this plan’s kernel.

pub fn run( &self, stream: &Stream, _workspace: Workspace<'_>, args: HistogramArgs<'_, T>, ) -> Result<()>

Launch.