Interface Tensor

Represent multi-dimensional arrays to feed to or fetch from model inferencing.

Hierarchy

TypedTensorBase<Type>
TypedTensorUtils<Type>
- Tensor

Index

Properties

data dims size type

Methods

reshape toImageData

Properties

`Readonly` data

data: string[] | Int8Array | Uint8Array | Int16Array | Uint16Array | Int32Array | Uint32Array | Float32Array | Float64Array | BigInt64Array | BigUint64Array

Get the buffer data of the tensor.

`Readonly` dims

dims: readonly number[]

Get the dimensions of the tensor.

`Readonly` size

size: number

Get the number of elements in the tensor.

`Readonly` type

type: keyof DataTypeMap

Get the data type of the tensor.

Methods

reshape

reshape(dims: readonly number[]): TypedTensor<keyof DataTypeMap>
Create a new tensor with the same data buffer and specified dims.
Parameters
- dims: readonly number[]
  
  New dimensions. Size should match the old one.
Returns TypedTensor<keyof DataTypeMap>
Inherited from TypedTensorUtils.reshape
- Defined in tensor-utils.ts:19

toImageData

toImageData(options?: TensorToImageDataOptions): ImageData
creates an ImageData instance from tensor

Returns
An ImageData instance which can be used to draw on canvas
Parameters
- Optional options: TensorToImageDataOptions
Returns ImageData
Inherited from TypedTensorUtils.toImageData
- Defined in tensor-utils.ts:31

Generated using TypeDoc