Struct UniformBatchedSpMV 

pub struct UniformBatchedSpMV<T> { /* private fields */ }

Optimised batched SpMV when all matrices share the same sparsity pattern.

Only the non-zero values differ between matrices; the row pointers and column indices are shared. This enables more efficient memory layout and potential vectorisation across the batch dimension.

Implementations

impl<T: GpuFloat + Add<Output = T> + Mul<Output = T> + AddAssign> UniformBatchedSpMV<T>

pub fn from_pattern( pattern: &CsrMatrix<T>, batch_values: Vec<Vec<T>>, ) -> SparseResult<Self>

Create a uniform batch from a pattern matrix and per-batch values.

The pattern matrix supplies the shared structure (row_ptr, col_idx); batch_values[i] supplies the non-zero values for the i-th matrix.

Errors

Returns SparseError::InvalidArgument if batch_values is empty or if any entry has the wrong length. Returns SparseError::Cuda on download failure.

pub fn from_host_arrays( rows: usize, cols: usize, row_ptr: Vec<i32>, col_idx: Vec<i32>, batch_values: Vec<Vec<T>>, ) -> SparseResult<Self>

Create a uniform batch from host-side arrays (no GPU required).

Errors

Returns SparseError::InvalidArgument on dimension / length mismatch.

pub fn batch_size(&self) -> usize

Number of matrices in the batch.

pub fn execute( &self, xs: &[Vec<T>], ys: &mut [Vec<T>], alpha: T, beta: T, ) -> SparseResult<()>

Execute batched SpMV on the host.

Errors

Returns SparseError::DimensionMismatch on vector size mismatch.

Trait Implementations

impl<T: Clone> Clone for UniformBatchedSpMV<T>

fn clone(&self) -> UniformBatchedSpMV<T>

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<T: Debug> Debug for UniformBatchedSpMV<T>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.