Mirrors
/
hf_text-generation-inference
mirror of https://github.com/huggingface/text-generation-inference.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fixing rocm gptq by using triton code too (renamed cuda into triton). (#2691)

This commit is contained in:
Nicolas Patry 2024-10-25 09:17:57 +02:00 committed by GitHub
parent 43df056eee
commit cece8635f8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
server/text_generation_server/layers/gptq/__init__.py
View File

@ -10,8 +10,8 @@ from text_generation_server.utils.weights import Weight, Weights, WeightsLoader
if SYSTEM == "ipex": if SYSTEM == "ipex":
from .ipex import QuantLinear from .ipex import QuantLinear
elif SYSTEM == "cuda": elif SYSTEM in {"cuda", "rocm"}:
from .cuda import QuantLinear from .triton import QuantLinear
@dataclass @dataclass

0
server/text_generation_server/layers/gptq/cuda.py → server/text_generation_server/layers/gptq/triton.py
View File