Struct Tensor 

pub struct Tensor {
    pub legs: Vec<Index>,
    pub data: Vec<Complex<f32>>,
}

A dense complex tensor with named legs.

legs[d] is the Index of dimension d; every leg has extent 2 (a qubit wire). Data is stored row-major in leg order: the flat offset of a multi-index (i_0, …, i_{r-1}) is Σ_d i_d · 2^{r-1-d}.

Fields

legs: Vec<Index>

Names of the legs, in storage order.

data: Vec<Complex<f32>>

Dense amplitudes, length 2^legs.len().

Implementations

impl Tensor

pub fn new(legs: Vec<Index>, data: Vec<Complex<f32>>) -> QuantumResult<Self>

Build a tensor from explicit legs and data, validating the length.

Errors

Returns QuantumError::DimensionMismatch if data.len() != 2^legs.len().

pub fn scalar(value: Complex<f32>) -> Self

Rank-0 scalar tensor.

pub fn ket_zero(wire: Index) -> Self

The single-qubit |0⟩ state vector as a rank-1 tensor on leg wire.

pub fn bra_basis(wire: Index, bit: bool) -> Self

The single-qubit basis bra ⟨b| as a rank-1 tensor on leg wire.

pub fn gate_1q( out_wire: Index, in_wire: Index, m: &[[Complex<f32>; 2]; 2], ) -> Self

A single-qubit gate tensor out·in with legs [out_wire, in_wire].

m[r][c] is the matrix entry ⟨r|G|c⟩; the out_wire leg carries the row index, the in_wire leg the column index.

pub fn gate_2q( out0: Index, out1: Index, in0: Index, in1: Index, m: &[[Complex<f32>; 4]; 4], ) -> Self

A two-qubit gate tensor with legs [out0, out1, in0, in1].

m is the 4 × 4 matrix in row-major form using the convention that the composite index is 2*q0 + q1 (qubit 0 most significant), i.e. the same ordering used by the dense two-qubit application path.

pub fn rank(&self) -> usize

Rank of the tensor (number of legs).

pub fn as_scalar(&self) -> QuantumResult<Complex<f32>>

Extract the scalar value of a rank-0 tensor.

Errors

Returns QuantumError::Internal if the tensor is not rank-0.

pub fn contract(&self, other: &Self) -> QuantumResult<Self>

Contract self with other over all shared legs (a tensordot).

The resulting tensor’s legs are self’s remaining legs followed by other’s remaining legs, both in their original order. Shared legs are summed over. With no shared legs this reduces to an outer product.

Errors

Currently infallible for well-formed inputs; returns a Result for interface uniformity and future fallibility.

Trait Implementations

impl Clone for Tensor

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Tensor

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

Formats the value using the given formatter.