root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-13 12:10:18 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

fix types

Browse Source
This commit is contained in:
ngxson 2024-07-10 00:36:52 +02:00
parent ee2b35c65f
commit 713665db2e
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
convert_lora_to_gguf.py
View File

@ -128,7 +128,7 @@ if __name__ == '__main__':
map_name[tensor_name] = dest_name map_name[tensor_name] = dest_name
# overwrite method # overwrite method
def map_tensor_name(self, name: str) -> Iterator[tuple[str, Tensor]]: def map_tensor_name(self, name: str) -> str:
return map_name[name] return map_name[name]
# overwrite method # overwrite method
@ -141,13 +141,13 @@ if __name__ == '__main__':
del name, new_name, bid, n_dims # unused del name, new_name, bid, n_dims # unused
return ftype != gguf.LlamaFileType.ALL_F32 return ftype != gguf.LlamaFileType.ALL_F32
model_instance._map_tensor_name = model_instance.map_tensor_name model_instance._map_tensor_name = model_instance.map_tensor_name # type: ignore
model_instance.map_tensor_name = types.MethodType(map_tensor_name, model_instance) model_instance.map_tensor_name = types.MethodType(map_tensor_name, model_instance)
model_instance._get_tensors = model_instance.get_tensors model_instance._get_tensors = model_instance.get_tensors # type: ignore
model_instance.get_tensors = types.MethodType(get_tensors, model_instance) model_instance.get_tensors = types.MethodType(get_tensors, model_instance)
model_instance._extra_f16_tensors = model_instance.extra_f16_tensors model_instance._extra_f16_tensors = model_instance.extra_f16_tensors # type: ignore
model_instance.extra_f16_tensors = types.MethodType(extra_f16_tensors, model_instance) model_instance.extra_f16_tensors = types.MethodType(extra_f16_tensors, model_instance)
model_instance.gguf_writer.add_quantization_version(gguf.GGML_QUANT_VERSION) model_instance.gguf_writer.add_quantization_version(gguf.GGML_QUANT_VERSION)