[Core]Refactor gptq_marlin ops (#4466)
This commit is contained in:
@@ -6,7 +6,7 @@ import numpy
|
||||
import torch
|
||||
from torch.nn.parameter import Parameter
|
||||
|
||||
from vllm._C import ops
|
||||
from vllm import _custom_ops as ops
|
||||
from vllm.model_executor.layers.linear import (LinearBase, LinearMethodBase,
|
||||
set_weight_attrs)
|
||||
from vllm.model_executor.layers.quantization.base_config import (
|
||||
|
||||
Reference in New Issue
Block a user