FlexDelegate class

Flex delegate for running models that use SELECT_TF_OPS.

The Flex delegate enables TensorFlow operations that are not available as TFLite builtins. This is required for training models whose gradient ops cannot be expressed as builtins (e.g., Conv2D, BatchNormalization).

Add flutter_litert_flex to your pubspec.yaml to bundle the native library on supported native platforms:

dependencies:
  flutter_litert: ^3.2.2
  flutter_litert_flex: ^1.3.0

Then use the async constructor, which is required on Android and also works on the other supported native platforms:

final flex = await FlexDelegate.create();
final options = InterpreterOptions();
options.addDelegate(flex);
final interpreter = Interpreter.fromFile(model, options: options);

Implemented types

• Delegate

Constructors

FlexDelegate()

Creates a FlexDelegate for SELECT_TF_OPS support.

factory

Properties

base → Pointer<TfLiteDelegate>

Get pointer to TfLiteDelegate

no setter, override

hashCode → int

The hash code for this object.

no setter, inherited

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

delete() → void

Releases native Flex delegate resources.

override

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited

Static Properties

isAvailable → bool

Whether the Flex delegate library is available.

no setter

Static Methods

create() → Future<FlexDelegate>

Creates a FlexDelegate asynchronously.