Function: tensor()

tensor(dtype, shape, src?): Tensor

Defined in: core/tensor.ts:147

Allocates a new native tensor with the specified data type and shape.

Optionally initializes the tensor's buffer from a typed array src. When src is omitted the buffer contents are undefined. The returned tensor resides in native C++ memory; call Tensor.dispose when the tensor is no longer needed.

Parameters

dtype

DType

The element data type of the tensor.

shape

number[]

An array of dimension sizes (e.g. [1, 3, 224, 224]).

src?

Optional typed array used to initialize the tensor's data. Its size in bytes must match tensor's size.

Int32Array<ArrayBufferLike> | Float32Array<ArrayBufferLike> | Uint8Array<ArrayBufferLike> | BigInt64Array<ArrayBufferLike>

Returns

Tensor

A newly allocated native tensor.

Throws

Thrown with code INVALID_ARGUMENT if any dimension in shape is non-positive or if src byte length does not match the allocated tensor size.

Example

const t = tensor('float32', [1, 4], new Float32Array([1.0, 2.0, 3.0, 4.0]));
try {
  const data = t.getData(new Float32Array(4));
  console.log(data); // Float32Array [1, 2, 3, 4]
} finally {
  t.dispose();
}