Linear quantization operator. It consumes a high precision tensor, a scale, and a zero point to compute the low precision / quantized tensor. The quantization formula is y = saturate ((x / y_scale) + y_zero). For saturation, it saturates to [0, 255] if it's uint8, or [-128, 127] if it's int8. For (x / y_scale), it's rounding to the nearest even. More...

Collaboration diagram for QuantizeLinear:

Modules
	QuantizeLinearKernels

Classes
class	QuantizeLinearGraph< QUANTIZE_LINEAR, TT, INP_H, INP_W, OUT_W >
	Single instance graph. More...

class	QuantizeLinearStreamGraph< QUANTIZE_LINEAR, TT, INP_H, INP_W, OUT_W >
	Single instance stream graph. More...

class	QuantizeLinearChunkHPktStreamGraph< QUANTIZE_LINEAR, HCHUNK, TT, INP_H, INP_W, OUT_W >
	Multi instance pktstream graph. More...

Detailed Description

Template Parameters

QUANTIZE_LINAER	QuantizeLinaer Kernel
INP_H	input height
INP_W	input width
OUT_W	output width, allows padding, expects OUT_W >= INP_W

Modules

Classes

Detailed Description