ExecuTorchPreprocessor<T> class
abstract
Abstract base class for input preprocessing
This class defines the contract for preprocessing input data before inference. Implementations should convert domain-specific input (images, text, audio) into tensors suitable for model execution.
Type parameter T represents the input data type (e.g., Uint8List
for images, String for text, Float32List for audio).
Constructors
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
- inputTypeName → String
-
Human-readable name for the input type this preprocessor handles
no setter
- 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
-
preprocess(
T input) → Future< List< TensorData> > - Preprocesses input data into tensors for model inference
-
toString(
) → String -
A string representation of this object.
inherited
-
validateInput(
T input) → bool - Validates that the input data is suitable for preprocessing
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited