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

Update llama.cpp

Browse source
This commit is contained in:
Andrei Betlen 2023-03-28 21:10:23 -04:00
parent ea41474e04
commit 8908f4614c
2 changed files with 11 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
llama_cpp/llama_cpp.py
View file

@ -3,7 +3,6 @@ import ctypes
from ctypes import ( from ctypes import (
c_int, c_int,
c_float, c_float,
c_double,
c_char_p, c_char_p,
c_void_p, c_void_p,
c_bool, c_bool,
@ -40,7 +39,7 @@ class llama_token_data(Structure):
llama_token_data_p = POINTER(llama_token_data) llama_token_data_p = POINTER(llama_token_data)
llama_progress_callback = ctypes.CFUNCTYPE(None, c_double, c_void_p) llama_progress_callback = ctypes.CFUNCTYPE(None, c_float, c_void_p)
class llama_context_params(Structure): class llama_context_params(Structure):
@ -48,6 +47,7 @@ class llama_context_params(Structure):
("n_ctx", c_int), # text context ("n_ctx", c_int), # text context
("n_parts", c_int), # -1 for default ("n_parts", c_int), # -1 for default
("seed", c_int), # RNG seed, 0 for random ("seed", c_int), # RNG seed, 0 for random
("f16_kv", c_bool), # use fp16 for KV cache ("f16_kv", c_bool), # use fp16 for KV cache
( (
"logits_all", "logits_all",
@ -56,6 +56,7 @@ class llama_context_params(Structure):
("vocab_only", c_bool), # only load the vocabulary, no weights ("vocab_only", c_bool), # only load the vocabulary, no weights
("use_mlock", c_bool), # force system to keep model in RAM ("use_mlock", c_bool), # force system to keep model in RAM
("embedding", c_bool), # embedding mode only ("embedding", c_bool), # embedding mode only
# called with a progress value between 0 and 1, pass NULL to disable # called with a progress value between 0 and 1, pass NULL to disable
("progress_callback", llama_progress_callback), ("progress_callback", llama_progress_callback),
# context pointer passed to the progress callback # context pointer passed to the progress callback
@ -70,8 +71,7 @@ llama_context_params_p = POINTER(llama_context_params)
def llama_context_default_params() -> llama_context_params: def llama_context_default_params() -> llama_context_params:
params = _lib.llama_context_default_params() return _lib.llama_context_default_params()
return params
_lib.llama_context_default_params.argtypes = [] _lib.llama_context_default_params.argtypes = []
@ -229,9 +229,9 @@ def llama_sample_top_p_top_k(
last_n_tokens_data: llama_token_p, last_n_tokens_data: llama_token_p,
last_n_tokens_size: c_int, last_n_tokens_size: c_int,
top_k: c_int, top_k: c_int,
top_p: c_double, top_p: c_float,
temp: c_double, temp: c_float,
repeat_penalty: c_double, repeat_penalty: c_float,
) -> llama_token: ) -> llama_token:
return _lib.llama_sample_top_p_top_k( return _lib.llama_sample_top_p_top_k(
ctx, last_n_tokens_data, last_n_tokens_size, top_k, top_p, temp, repeat_penalty ctx, last_n_tokens_data, last_n_tokens_size, top_k, top_p, temp, repeat_penalty
@ -243,9 +243,9 @@ _lib.llama_sample_top_p_top_k.argtypes = [
llama_token_p, llama_token_p,
c_int, c_int,
c_int, c_int,
c_double, c_float,
c_double, c_float,
c_double, c_float,
] ]
_lib.llama_sample_top_p_top_k.restype = llama_token _lib.llama_sample_top_p_top_k.restype = llama_token

2
vendor/llama.cpp vendored

@ -1 +1 @@
Subproject commit 7e5395575a3360598f2565c73c8a2ec0c0abbdb8 Subproject commit 5a5f8b1501fbb34367225544010ddfc306d6d2fe