Tensor

This page reflects zetic_mlange 1.8.1.

Tensor is the Dart data container used for model inputs and outputs. It stores bytes, a DataType, and a shape.

Import

import 'package:zetic_mlange/zetic_mlange.dart';

Constructors

`Tensor`

Tensor({
  required Uint8List data,
  DataType dataType = DataType.int8,
  List<int>? shape,
})

Creates a tensor with defensive-copy semantics. If shape is omitted, the tensor shape defaults to [data.lengthInBytes].

`Tensor.view`

Tensor.view({
  required Uint8List data,
  DataType dataType = DataType.int8,
  List<int>? shape,
})

Creates a tensor view over the provided byte list without copying. Mutating the source Uint8List also mutates the tensor bytes.

For every constructor and factory, shape must describe exactly data.lengthInBytes / dataType.bytesPerElement elements. A mismatch throws MlangeException.

Typed Factories

Tensor.float32List(Float32List data, {List<int>? shape})
Tensor.float32View(Float32List data, {List<int>? shape})
Tensor.float64List(Float64List data, {List<int>? shape})
Tensor.float64View(Float64List data, {List<int>? shape})
Tensor.int32List(Int32List data, {List<int>? shape})
Tensor.int32View(Int32List data, {List<int>? shape})
Tensor.int64List(Int64List data, {List<int>? shape})
Tensor.int64View(Int64List data, {List<int>? shape})
Tensor.bytes(List<int> data, {DataType dataType = DataType.int8, List<int>? shape})
Tensor.bytesView(Uint8List data, {DataType dataType = DataType.int8, List<int>? shape})
Tensor.random(DataType dataType, List<int> shape, {Random? random})

Use List factories when you want Dart to copy the source data. Use View factories when you already own a correctly laid-out typed list and want to avoid an extra Dart-side allocation.

Properties

Property	Type	Description
`data`	`Uint8List`	Raw tensor bytes.
`dataType`	`DataType`	Tensor element data type.
`shape`	`List<int>`	Tensor shape.
`byteLength`	`int`	Number of bytes in `data`.
`elementCount`	`int`	Number of elements implied by `shape`.

Methods

int count()
int size()
Uint8List asUint8List()
Float32List asFloat32List()
Float64List asFloat64List()
Int32List asInt32List()
Int64List asInt64List()

Typed accessors validate that dataType matches the requested view.

Method	Description
`count()`	Number of elements based on `data.lengthInBytes ~/ dataType.bytesPerElement`.
`size()`	Number of bytes in `data`.
`asUint8List()`	Raw byte view without dtype validation.
`asFloat32List()`	Typed `Float32List` view. Requires `DataType.float32`.
`asFloat64List()`	Typed `Float64List` view. Requires `DataType.float64`.
`asInt32List()`	Typed `Int32List` view. Requires `DataType.int32`.
`asInt64List()`	Typed `Int64List` view. Requires `DataType.int64`.

Example

final input = Tensor.float32View(
  Float32List.fromList([1.0, 2.0, 3.0, 4.0]),
  shape: const [1, 4],
);

print(input.count()); // 4
print(input.asFloat32List());

Tensor

On this page