Struct SparseTensor 

pub struct SparseTensor {
    pub values: Tensor,
    pub indices: Tensor,
    pub shape: Vec<usize>,
    pub ndim: usize,
    pub nnz: usize,
    pub is_coalesced: bool,
}

Sparse tensor representation using COO (coordinate) format

Fields

values: Tensor

Tensor values (non-zero elements)

indices: Tensor

Indices of non-zero elements [2, nnz] for 2D, [3, nnz] for 3D, etc.

shape: Vec<usize>

Shape of the full tensor

ndim: usize

Number of dimensions

nnz: usize

Number of non-zero elements

is_coalesced: bool

Whether the tensor is coalesced (indices are ordered and unique)

Implementations

impl SparseTensor

pub fn new( values: Tensor, indices: Tensor, shape: Vec<usize>, ) -> TorshResult<Self>

Create a new sparse tensor from values and indices

pub fn from_dense(dense: &Tensor) -> TorshResult<Self>

Create a sparse tensor from dense tensor

pub fn to_dense(&self) -> TorshResult<Tensor>

Convert sparse tensor to dense tensor

pub fn coalesce(&mut self) -> TorshResult<()>

Coalesce the sparse tensor (combine duplicate indices)

pub fn nnz(&self) -> usize

Get the number of non-zero elements

pub fn shape(&self) -> &[usize]

Get the shape of the tensor

pub fn ndim(&self) -> usize

Get the number of dimensions

pub fn is_coalesced(&self) -> bool

Check if the tensor is coalesced

Trait Implementations

impl Clone for SparseTensor

fn clone(&self) -> SparseTensor

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SparseTensor

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

Formats the value using the given formatter.