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

gguf : general usability improvements (#3409)

Browse source
This commit is contained in:
cebtenzzre 2023-10-02 14:58:46 -04:00 • committed by GitHub
parent 9476b01226
commit 0fe321031a
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 120 additions and 101 deletions
Download patch file Download diff file Expand all files Collapse all files

5
convert.py
View file

@ -41,8 +41,7 @@ if hasattr(faulthandler, 'register') and hasattr(signal, 'SIGUSR1'):
NDArray: TypeAlias = 'np.ndarray[Any, Any]'
ARCH=gguf.MODEL_ARCH.LLAMA
NAMES=gguf.MODEL_TENSOR_NAMES[ARCH]
ARCH = gguf.MODEL_ARCH.LLAMA
DEFAULT_CONCURRENCY = 8
#
@ -953,7 +952,7 @@ class OutputFile:
of.close()
def pick_output_type(model: LazyModel, output_type_str: str | None) -> GGMLFileType:
wq_type = model[NAMES[gguf.MODEL_TENSOR.ATTN_Q].format(bid=0)+".weight"].data_type
wq_type = model[gguf.TENSOR_NAMES[gguf.MODEL_TENSOR.ATTN_Q].format(bid=0)+".weight"].data_type
if output_type_str == "f32" or (output_type_str is None and wq_type == DT_F32):
return GGMLFileType.AllF32