FilesExpand file tree

 main

/

llava_cpp.py

Copy path

Blame

More file actions

Blame

More file actions

Latest commit

 

History

History

History

163 lines (133 loc) · 6.47 KB

 main

/

llava_cpp.py

Copy path

Top

File metadata and controls

• Code
• Blame

163 lines (133 loc) · 6.47 KB

Raw

Copy raw file

Download raw file

Open symbols panel

Edit and raw actions

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

import sys

import os

import ctypes

from ctypes import (

c_bool,

c_char_p,

c_int,

c_int8,

c_int32,

c_uint8,

c_uint32,

c_size_t,

c_float,

c_double,

c_void_p,

POINTER,

_Pointer, # type: ignore

Structure,

Array,

)

import pathlib

from typing import List, Union

import llama_cpp.llama_cpp as llama_cpp

# Load the library

def _load_shared_library(lib_base_name: str):

# Construct the paths to the possible shared library names

_base_path = pathlib.Path(os.path.abspath(os.path.dirname(__file__)))

# Searching for the library in the current directory under the name "libllama" (default name

# for llamacpp) and "llama" (default name for this repo)

_lib_paths: List[pathlib.Path] = []

# Determine the file extension based on the platform

if sys.platform.startswith("linux"):

_lib_paths += [

_base_path / f"lib{lib_base_name}.so",

]

elif sys.platform == "darwin":

_lib_paths += [

_base_path / f"lib{lib_base_name}.so",

_base_path / f"lib{lib_base_name}.dylib",

]

elif sys.platform == "win32":

_lib_paths += [

_base_path / f"{lib_base_name}.dll",

_base_path / f"lib{lib_base_name}.dll",

]

else:

raise RuntimeError("Unsupported platform")

if "LLAVA_CPP_LIB" in os.environ:

lib_base_name = os.environ["LLAVA_CPP_LIB"]

_lib = pathlib.Path(lib_base_name)

_base_path = _lib.parent.resolve()

_lib_paths = [_lib.resolve()]

cdll_args = dict() # type: ignore

# Add the library directory to the DLL search path on Windows (if needed)

if sys.platform == "win32" and sys.version_info >= (3, 8):

os.add_dll_directory(str(_base_path))

if "CUDA_PATH" in os.environ:

os.add_dll_directory(os.path.join(os.environ["CUDA_PATH"], "bin"))

os.add_dll_directory(os.path.join(os.environ["CUDA_PATH"], "lib"))

cdll_args["winmode"] = ctypes.RTLD_GLOBAL

# Try to load the shared library, handling potential errors

for _lib_path in _lib_paths:

if _lib_path.exists():

try:

return ctypes.CDLL(str(_lib_path), **cdll_args)

except Exception as e:

raise RuntimeError(f"Failed to load shared library '{_lib_path}': {e}")

raise FileNotFoundError(

f"Shared library with base name '{lib_base_name}' not found"

)

# Specify the base name of the shared library to load

_libllava_base_name = "llava"

# Load the library

_libllava = _load_shared_library(_libllava_base_name)

################################################

# llava.h

################################################

# struct clip_ctx;

clip_ctx_p = c_void_p

# struct llava_image_embed {

# float * embed;

# int n_image_pos;

# };

class llava_image_embed(Structure):

_fields_ = [

("embed", POINTER(c_float)),

("n_image_pos", c_int),

]

# /** sanity check for clip <-> llava embed size match */

# LLAVA_API bool llava_validate_embed_size(const llama_context * ctx_llama, const clip_ctx * ctx_clip);

def llava_validate_embed_size(ctx_llama: llama_cpp.llama_context_p, ctx_clip: clip_ctx_p) -> bool:

return _libllava.llava_validate_embed_size(ctx_llama, ctx_clip)

_libllava.llava_validate_embed_size.argtypes = [llama_cpp.llama_context_p, clip_ctx_p]

_libllava.llava_validate_embed_size.restype = c_bool

# /** build an image embed from image file bytes */

# LLAVA_API struct llava_image_embed * llava_image_embed_make_with_bytes(struct clip_ctx * ctx_clip, int n_threads, const unsigned char * image_bytes, int image_bytes_length);

def llava_image_embed_make_with_bytes(ctx_clip: clip_ctx_p, n_threads: Union[c_int, int], image_bytes: bytes, image_bytes_length: Union[c_int, int]) -> "_Pointer[llava_image_embed]":

return _libllava.llava_image_embed_make_with_bytes(ctx_clip, n_threads, image_bytes, image_bytes_length)

_libllava.llava_image_embed_make_with_bytes.argtypes = [clip_ctx_p, c_int, POINTER(c_uint8), c_int]

_libllava.llava_image_embed_make_with_bytes.restype = POINTER(llava_image_embed)

# /** build an image embed from a path to an image filename */

# LLAVA_API struct llava_image_embed * llava_image_embed_make_with_filename(struct clip_ctx * ctx_clip, int n_threads, const char * image_path);

def llava_image_embed_make_with_filename(ctx_clip: clip_ctx_p, n_threads: Union[c_int, int], image_path: bytes) -> "_Pointer[llava_image_embed]":

return _libllava.llava_image_embed_make_with_filename(ctx_clip, n_threads, image_path)

_libllava.llava_image_embed_make_with_filename.argtypes = [clip_ctx_p, c_int, c_char_p]

_libllava.llava_image_embed_make_with_filename.restype = POINTER(llava_image_embed)

# LLAVA_API void llava_image_embed_free(struct llava_image_embed * embed);

# /** free an embedding made with llava_image_embed_make_* */

def llava_image_embed_free(embed: "_Pointer[llava_image_embed]"):

return _libllava.llava_image_embed_free(embed)

_libllava.llava_image_embed_free.argtypes = [POINTER(llava_image_embed)]

_libllava.llava_image_embed_free.restype = None

# /** write the image represented by embed into the llama context with batch size n_batch, starting at context pos n_past. on completion, n_past points to the next position in the context after the image embed. */

# LLAVA_API bool llava_eval_image_embed(struct llama_context * ctx_llama, const struct llava_image_embed * embed, int n_batch, int * n_past);

def llava_eval_image_embed(ctx_llama: llama_cpp.llama_context_p, embed: "_Pointer[llava_image_embed]", n_batch: Union[c_int, int], n_past: "_Pointer[c_int]") -> bool:

return _libllava.llava_eval_image_embed(ctx_llama, embed, n_batch, n_past)

_libllava.llava_eval_image_embed.argtypes = [llama_cpp.llama_context_p, POINTER(llava_image_embed), c_int, POINTER(c_int)]

_libllava.llava_eval_image_embed.restype = c_bool

################################################

# clip.h

################################################

# /** load mmproj model */

# CLIP_API struct clip_ctx * clip_model_load (const char * fname, int verbosity);

def clip_model_load(fname: bytes, verbosity: Union[c_int, int]) -> clip_ctx_p:

return _libllava.clip_model_load(fname, verbosity)

_libllava.clip_model_load.argtypes = [c_char_p, c_int]

_libllava.clip_model_load.restype = clip_ctx_p

# /** free mmproj model */

# CLIP_API void clip_free(struct clip_ctx * ctx);

def clip_free(ctx: clip_ctx_p):

return _libllava.clip_free(ctx)

_libllava.clip_free.argtypes = [clip_ctx_p]

_libllava.clip_free.restype = None