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

gguf-py : fix flake8 lint

Browse source
This commit is contained in:
Francis Couture-Harpin 2024-08-02 16:23:30 -04:00
parent 5e27e7e11c
commit 861265b91e
2 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

1
gguf-py/gguf/constants.py
View file

@ -1236,6 +1236,7 @@ LlamaFileTypeMap: dict[LlamaFileType, GGMLQuantizationType] = {
LlamaFileType.MOSTLY_Q4_0_8_8: GGMLQuantizationType.Q4_0_8_8,
}
class GGUFEndian(IntEnum):
LITTLE = 0
BIG = 1

4
gguf-py/gguf/quants.py
View file

@ -69,12 +69,12 @@ class __Quant(ABC):
@classmethod
@abstractmethod
def quantize_blocks(cls, blocks: np.ndarray) -> np.ndarray:
raise NotImplemented
raise NotImplementedError
@classmethod
@abstractmethod
def dequantize_blocks(cls, blocks: np.ndarray) -> np.ndarray:
raise NotImplemented
raise NotImplementedError
@classmethod
def quantize_rows(cls, rows: np.ndarray) -> np.ndarray: