mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-28 12:24:35 +00:00
20 lines
642 B
C
20 lines
642 B
C
|
#pragma once
|
||
|
|
|
||
|
|
#include "ggml.h"
|
||
|
|
|
||
|
|
// GGML internal header
|
||
|
|
|
||
|
|
#ifdef __cplusplus
|
||
|
|
extern "C" {
|
||
|
|
#endif
|
||
|
|
|
||
|
|
// Quantization utilizing an importance matrix (a.k.a. "Activation aWare Quantization")
|
||
|
|
size_t quantize_q4_0_4x4(const float * GGML_RESTRICT src, void * GGML_RESTRICT dst, int64_t nrows, int64_t n_per_row, const float * imatrix);
|
||
|
|
size_t quantize_q4_0_4x8(const float * GGML_RESTRICT src, void * GGML_RESTRICT dst, int64_t nrows, int64_t n_per_row, const float * imatrix);
|
||
|
|
size_t quantize_q4_0_8x8(const float * GGML_RESTRICT src, void * GGML_RESTRICT dst, int64_t nrows, int64_t n_per_row, const float * imatrix);
|
||
|
|
|
||
|
|
#ifdef __cplusplus
|
||
|
|
}
|
||
|
|
#endif
|
||
|
|
|