root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-11-11 13:30:35 +00:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

Comment explaining a decision (#6531)

Browse Source
This commit is contained in:
kunnis 2024-04-08 10:44:19 -05:00 committed by GitHub
parent b73e564b16
commit cecd8d3c98
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
convert.py
View File

@ -139,7 +139,8 @@ class GGMLFileType(enum.IntEnum):
dt = GGML_FILE_TYPE_TO_DATA_TYPE.get(self)
if dt is None:
raise ValueError(self)
# 1D tensors are always F32.
# Convert all 1D tensors to F32. Most of the codebase that takes in 1D tensors only handles F32 tensors, and most of the outputs tensors are F32.
# Also The 1d tensors aren't much of a performance/size issue. So instead of having to have separate F32 and F16 implementations of both, just convert everything to F32 for now.
return dt if len(tensor.shape) > 1 else DT_F32