DequantizeLinear

Linear dequantization operator. It consumes a quantized tensor, a scale, and a zero point to compute the full precision tensor. The dequantization formula is y = (x - x_zero_point) * x_scale. x_scale and x_zero_point must have same shape, and can be either a scalar for per-tensor / per layer quantization, or a 1-D tensor for per-axis quantization. x_zero_point and x must have same type. x and y must have same shape. More...

Collaboration diagram for DequantizeLinear:

Modules
	DequantizeLinearKernels

Classes
class	DequantizeLinearGraph< DEQUANTIZE_LINEAR, TT, B, INP_W, OUT_W >
	Single instance graph. More...

Detailed Description

Template Parameters

DEQUANTIZE_LINEAR	DequantizeLinear Kernel
TT	int8_t or uint8_t
B	batch size
INP_W	input width
OUT_W	output width, expect OUT_W > INP_W