vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Make the GGML header look nicer

Browse source
This commit is contained in:
Justine Tunney 2024-04-21 13:39:40 -07:00
parent d6892c486b
commit ce0442d7a7
No known key found for this signature in database
GPG key ID: 52965314629936D4
1 changed files with 8 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

18
ggml.h
View file

@ -326,18 +326,16 @@ extern "C" {
// get ggml_status name string // get ggml_status name string
GGML_API GGML_CALL const char * ggml_status_to_string(enum ggml_status status); GGML_API GGML_CALL const char * ggml_status_to_string(enum ggml_status status);
// ieee 754-2008 half-precision float16
// todo: make this not an integral type
typedef uint16_t ggml_fp16_t; typedef uint16_t ggml_fp16_t;
GGML_API float ggml_fp16_to_fp32(ggml_fp16_t);
GGML_API ggml_fp16_t ggml_fp32_to_fp16(float);
GGML_API void ggml_fp16_to_fp32_row(const ggml_fp16_t *, float *, int64_t);
GGML_API void ggml_fp32_to_fp16_row(const float *, ggml_fp16_t *, int64_t);
// convert FP16 <-> FP32 // google brain half-precision bfloat16
GGML_API float ggml_fp16_to_fp32(ggml_fp16_t x); typedef struct { uint16_t bits; } ggml_bf16_t;
GGML_API ggml_fp16_t ggml_fp32_to_fp16(float x);
GGML_API void ggml_fp16_to_fp32_row(const ggml_fp16_t * x, float * y, int64_t n);
GGML_API void ggml_fp32_to_fp16_row(const float * x, ggml_fp16_t * y, int64_t n);
// bfloat16
struct ggml_bf16_s;
typedef struct ggml_bf16_s ggml_bf16_t;
GGML_API ggml_bf16_t ggml_fp32_to_bf16(float); GGML_API ggml_bf16_t ggml_fp32_to_bf16(float);
GGML_API float ggml_bf16_to_fp32(ggml_bf16_t); // consider just doing << 16 GGML_API float ggml_bf16_to_fp32(ggml_bf16_t); // consider just doing << 16
GGML_API void ggml_bf16_to_fp32_row(const ggml_bf16_t *, float *, int64_t); GGML_API void ggml_bf16_to_fp32_row(const ggml_bf16_t *, float *, int64_t);