ExecuTorchAdapter class

Runs executorch variants (.pte programs) through executorch_flutter.

ExecuTorch takes inputs by position, so they are passed in the order the manifest lists them, and outputs are matched to manifest outputs by order.

Constructors

ExecuTorchAdapter({Set<String>? backends})

Properties

backends → Set<String>

Variant backends this adapter accepts. A variant without a backend is always accepted.

final

hashCode → int

The hash code for this object.

no setter, inherited

runtime → String

The manifest runtime this adapter runs, such as onnx. See Runtimes.

no setter

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

canRun(Variant variant) → bool

Whether this adapter can run variant on this device, for example whether it supports the variant's backend.

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

open(LoadedVariant model) → Future<TensorSession>

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited

Static Methods

defaultBackends() → Set<String>

XNNPACK everywhere, plus CoreML and MPS on Apple platforms.