multiply

Function Usage

Image processing algorithms, tensor multiply algorithm, supporting dtype.float16, dtype.float32 and dtype.uint8.

It is supported by the Atlas inference product and Atlas 200I/500 A2 inference product.

  • The input and output tensors must be on the device.
  • The input parameters cannot exceed four dimensions, and must match the tensor shapes.
  • If scale is not passed, the data types of the input parameters must be the same. If scale is passed, different input data types are supported. The output data type is the same as that of the input tensor with high precision.
  • Handle the issue of out-of-range data type if any.

Prototype

multiply(inputTensor1: Tensor, inputTensor2: Tensor, scale)

Input Parameters

Parameter

Type

Description

inputTensor1

Tensor class

Multiplier, input tensor, supporting dtype.float16, dtype.float32, and dtype.uint8 data types.

inputTensor2

Tensor class

Multiplier, input tensor, supporting dtype.float16, dtype.float32, and dtype.uint8 data types.

scale

float

(Optional) Multiplier for scaling the result.

Response Parameters

Tensor data after the multiplication