alihan/llama-cpp-python
1
0
Fork 0
mirror of https://github.com/abetlen/llama-cpp-python.git synced 2023-09-07 17:34:22 +03:00
Code Issues Packages Projects Releases Wiki Activity

Update llama.cpp

Browse Source
This commit is contained in:
Andrei Betlen
2023-03-25 16:26:03 -04:00
parent b121b7c05b
commit 4525236214
2 changed files with 10 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
llama_cpp/llama_cpp.py
View File

@@ -42,6 +42,7 @@ 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_double, c_void_p)
class llama_context_params(Structure): class llama_context_params(Structure):
_fields_ = [ _fields_ = [
("n_ctx", c_int), # text context ("n_ctx", c_int), # text context
@@ -55,7 +56,6 @@ 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
@@ -165,6 +165,14 @@ _lib.llama_n_ctx.argtypes = [llama_context_p]
_lib.llama_n_ctx.restype = c_int _lib.llama_n_ctx.restype = c_int
def llama_n_embd(ctx: llama_context_p) -> c_int:
return _lib.llama_n_ctx(ctx)
_lib.llama_n_embd.argtypes = [llama_context_p]
_lib.llama_n_embd.restype = c_int
# Token logits obtained from the last call to llama_eval() # Token logits obtained from the last call to llama_eval()
# The logits for the last token are stored in the last row # The logits for the last token are stored in the last row
# Can be mutated in order to change the probabilities of the next token # Can be mutated in order to change the probabilities of the next token

2
vendor/llama.cpp vendored

Submodule vendor/llama.cpp updated: 502a400192...c2b25b6912