[object Object][object Object][object Object]undefined
[object Object]

Performs element-wise quantization on the input int32_t type data and converts it to half type. The computation formula is as follows:

dst[i]=Cast(src[i]scale)dst[i] = Cast(src[i] * scale)

where scale can be set through .

[object Object]
  • First-n data computation

    [object Object]
  • High-dimensional tiling computation

    [object Object]
  • Synchronous computation

    [object Object]
[object Object][object Object]undefined
[object Object]

None

[object Object]

PIPE_V

[object Object]
  • Must be used together with the non-vector mode of the interface, which is used to set the quantization parameters.
  • The starting addresses of dst and src must be 32-byte aligned.
  • For operand address overlap constraints, refer to .
[object Object]
[object Object]