root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-24 10:24:35 +00:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

convert : fix conversion of some BERT embedding models (#6937)

Browse Source
This commit is contained in:
Christian Zhou-Zheng 2024-04-29 09:34:41 -04:00 committed by GitHub
parent 577277ffd2
commit 3055a41805
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
convert-hf-to-gguf.py
View File

@ -2482,6 +2482,10 @@ class BertModel(Model):
print(f"Can not map tensor {name!r}")
sys.exit()
# convert any unsupported data types to float32
if data_torch.dtype not in (torch.float16, torch.float32):
data_torch = data_torch.to(torch.float32)
data = data_torch.squeeze().numpy()
n_dims = len(data.shape)
new_dtype: type[np.floating[Any]]