TensorUtils class
Utility class for working with ExecuTorch tensors
These helpers build TensorData directly rather than going through
ExecutorchManager.instance. Routing through the singleton would make them
throw on any platform without a native manager — notably web, where
package:executorch_flutter supplies the manager and this library's
ExecutorchManager.instance resolves to a stub that throws. Nothing here
needs a manager: the conversion is pure Dart.
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
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 Methods
-
calculateElementCount(
List< int> shape) → int - Calculate the total number of elements in a tensor shape
-
createFloat32Tensor2D(
{required List< List< data, String? name}) → TensorDatadouble> > - Create a float32 tensor from a 2D list (commonly used for images)
-
createFloat32Tensor3D(
{required List< List< data, String? name}) → TensorDataList< >double> > - Create a float32 tensor from a 3D list (commonly used for RGB images)
-
createFloat32Tensor4D(
{required List< List< data, String? name}) → TensorDataList< >List< >double> > - Create a float32 tensor from a 4D list (commonly used for batched images)
-
extractFloat32Data(
TensorData tensor) → List< double> - Extract numeric data from a tensor
-
extractInt32Data(
TensorData tensor) → List< int> - Extract integer data from a tensor
-
formatShape(
List< int> shape) → String - Format tensor shape as a human-readable string