Update quantization
This commit is contained in:
parent
c3dece3f01
commit
0101afdfa7
|
@ -7,7 +7,11 @@ import base64
|
||||||
import ctypes
|
import ctypes
|
||||||
|
|
||||||
from typing import List
|
from typing import List
|
||||||
|
|
||||||
|
try:
|
||||||
from cpm_kernels.kernels.base import LazyKernelCModule, KernelFunction, round_up
|
from cpm_kernels.kernels.base import LazyKernelCModule, KernelFunction, round_up
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
|
||||||
|
|
||||||
class W8A16Linear(torch.autograd.Function):
|
class W8A16Linear(torch.autograd.Function):
|
||||||
|
|
Loading…
Reference in New Issue