biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc] Update GPTQ to use vLLMParameters (#7976)

Browse Source
This commit is contained in:
Dipika Sikka
2024-09-03 17:21:44 -04:00
committed by GitHub
parent dc0b6066ab
commit 2188a60c7e
6 changed files with 93 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/model_executor/parameter.py
View File

@@ -1,3 +1,4 @@
from fractions import Fraction
from typing import Callable, Optional, Union
import torch
@@ -257,7 +258,7 @@ class PackedColumnParameter(_ColumnvLLMParameter):
"""
def __init__(self,
packed_factor: int,
packed_factor: Union[int, Fraction],
packed_dim: int,
marlin_tile_size: Optional[int] = None,
**kwargs):
@@ -298,7 +299,7 @@ class PackedvLLMParameter(ModelWeightParameter):
"""
def __init__(self,
packed_factor: int,
packed_factor: Union[int, Fraction],
packed_dim: int,
marlin_tile_size: Optional[int] = None,
**kwargs):