Vector implementation for MK*KN, stores weights and biases, requires N%16=0 Qgemm<a,a,1,80,32> takes cycles.

#include <qgemm.h>

Public Member Functions
	QgemmStream (int32_t(&b)[N], float x_scale, float w_scale, float y_scale, TT x_zero, TTPARAM w_zero, TT y_zero)

void	filter (input_stream< TT > in, input_stream< TTPARAM > weight, output_stream< TT > *out)

Static Public Member Functions
static void	registerKernelClass ()

The documentation for this class was generated from the following files:

Public Member Functions

Static Public Member Functions