Trait EagerEinsumExt

Source

pub trait EagerEinsumExt {
    // Required methods
    fn einsum(&self, subscripts: &str) -> Result<EagerTensor>;
    fn einsum_notation(&self, notation: &EinsumNotation) -> Result<EagerTensor>;
    fn einsum_subscripts(
        &self,
        subscripts: &EinsumSubscripts,
    ) -> Result<EagerTensor>;
}

Expand description

Eager einsum extension methods for slices or arrays of EagerTensor refs.

Required Methods§

Source

fn einsum(&self, subscripts: &str) -> Result<EagerTensor>

Execute an einsum from string notation.

§Errors

Returns Error::InvalidSubscripts for malformed notation, Error::Validation for rank/shape/dtype mismatches, or Error::Planning / Error::Runtime for contraction planning and execution failures.

Source

fn einsum_notation(&self, notation: &EinsumNotation) -> Result<EagerTensor>

Execute an einsum from rank-unresolved notation.

§Examples

use tenferro_einsum::{EinsumAxis, EinsumNotation};
let notation = EinsumNotation::new(&[&[EinsumAxis::Ellipsis]], &[]);
assert_eq!(notation.input_count(), 1);

§Errors

Returns a typed validation, planning, or runtime error when notation or execution is invalid.

Source

fn einsum_subscripts( &self, subscripts: &EinsumSubscripts, ) -> Result<EagerTensor>

Execute an einsum from parsed integer labels.

§Errors

Returns Error::Validation for rank/shape/dtype mismatches, Error::Planning for an invalid contraction plan, or Error::Runtime for extension registration or backend execution failures.