Struct Df64EinsumVjp 

pub struct Df64EinsumVjp { /* private fields */ }

The adjoint of a two-input contraction.

The adjoint of out = einsum(lhs, rhs) contracts the output cotangent with the other operand, which is the same operation with the labels rotated: lhs_bar = einsum(out, rhs -> lhs) and rhs_bar = einsum(lhs, out -> rhs). It carries the pattern so the adjoint uses exactly the labels the primal used, and it has two outputs because the contraction has two inputs.

Examples

use tenferro_ad::extension::ExtensionOp;
use tenferro_df64_proof::extension::Df64EinsumVjp;

let adjoint = Df64EinsumVjp::of(&[&[0, 1], &[1, 2]], &[0, 2]).expect("a contraction");
assert_eq!(<Df64EinsumVjp as ExtensionOp>::input_count(&adjoint), 3);
assert_eq!(<Df64EinsumVjp as ExtensionOp>::output_count(&adjoint), 2);

Implementations

impl Df64EinsumVjp

pub fn of(inputs: &[&[u32]], out: &[u32]) -> Result<Self>

Build the adjoint for the same labels as the primal contraction.

Errors

Returns [tenferro_tensor::Error::InvalidArgument] when the pattern is not a pairwise contraction: an operand carries no label, a label repeats inside one operand, the operands share no contracted label, or an output label appears in no operand.

Examples

use tenferro_df64_proof::extension::Df64EinsumVjp;

assert!(Df64EinsumVjp::of(&[&[0, 1], &[1, 2]], &[0, 2]).is_ok());
assert!(Df64EinsumVjp::of(&[&[], &[1, 2]], &[0, 2]).is_err());

pub fn input_labels(&self) -> &[Vec<u32>]

The primal pattern’s label lists.

Every operand’s labels, in operand order.

Examples

use tenferro_df64_proof::extension::Df64EinsumVjp;

let adjoint = Df64EinsumVjp::of(&[&[0, 1], &[1, 2]], &[0, 2]).expect("a contraction");
assert_eq!(adjoint.input_labels(), &[vec![0, 1], vec![1, 2]]);

pub fn out_labels(&self) -> &[u32]

The output’s labels, in the output’s axis order.

Examples

use tenferro_df64_proof::extension::Df64EinsumVjp;

let adjoint = Df64EinsumVjp::of(&[&[0, 1], &[1, 2]], &[0, 2]).expect("a contraction");
assert_eq!(adjoint.out_labels(), &[0, 2]);

Trait Implementations

impl Clone for Df64EinsumVjp

fn clone(&self) -> Df64EinsumVjp

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Df64EinsumVjp

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

Formats the value using the given formatter.

impl Eq for Df64EinsumVjp

impl ExtensionOp for Df64EinsumVjp

fn family_id(&self) -> &'static str

Stable, process-independent family identifier.

fn payload_hash(&self, hasher: &mut dyn Hasher)

Hash the payload (everything except family_id).

fn payload_eq(&self, other: &dyn ExtensionOp) -> bool

Structural equality against another extension value.

fn clone_arc(&self) -> Arc<dyn ExtensionOp>

Deep-clone the payload behind an Arc.

fn as_any(&self) -> &dyn Any

Upcast this extension to &dyn Any for downcasting in payload_eq.

fn input_count(&self) -> usize

Number of primal inputs. MUST be constant for any given Arc<dyn ExtensionOp> value.

fn output_count(&self) -> usize

Number of outputs. MUST match the length of the vector returned by a successful Self::infer_output_meta call.

fn semantic_effects(&self) -> ExtensionEffectDeclaration<'_>

Declare observable semantic effects for this extension payload.

fn semantic_aliases(&self) -> ExtensionAliasDeclaration<'_>

Declare semantic output aliasing for this extension payload.

fn scalar_identity(&self) -> Option<&'static str>

Canonical identity of an externally defined scalar this operation carries.

fn infer_output_meta( &self, ctx: &mut ExtensionShapeContext<'_>, ) -> Result<Vec<(DType, Vec<SymDim>)>>

Infer output dtypes and shapes for each output slot.

fn lower_to_standard_ops( &self, _builder: &mut GraphBuilder<StdTensorOp>, _inputs: &[ValueRef<StdTensorOp>], _input_dtypes: &[DType], _input_shapes: &[&[SymDim]], ) -> Result<ExtensionStandardLowering, ExtensionLoweringError>

Try to expand this extension into standard tensor graph operations.

fn prune_outputs(&self, _live_outputs: &[bool]) -> Option<Arc<dyn ExtensionOp>>

Optionally return an equivalent op that produces only live outputs.

impl PartialEq for Df64EinsumVjp

fn eq(&self, other: &Df64EinsumVjp) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for Df64EinsumVjp