onnxruntime
Refine `op_types_to_quantize` argument handling in matmul_4bits_quantizer.py
#21815
Merged
Merged
Refine `op_types_to_quantize` argument handling in matmul_4bits_quantizer.py #21815
fajin-corp
merged
5 commits into
microsoft:main
from
duanshengliu:set-default-op-types-to-quantize
Set default op types to quantize as a list ["MatMul"]
db38eeea
Merge branch 'microsoft:main' into set-default-op-types-to-quantize
c6369e3b
fajin-corp
commented
on 2024-08-21
Merge branch 'microsoft:main' into set-default-op-types-to-quantize
36b576ed
update op_types_to_quantize
7e4e0260
duanshengliu
changed the title
Merge branch 'microsoft:main' into set-default-op-types-to-quantize
93dd4bd4
fajin-corp
approved these changes
on 2024-08-23
Login to write a write a comment.
Login via GitHub