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

fix: fix triton OutOfResources import

This commit is contained in:
OlivierDehaene 2023-12-14 16:04:26 +01:00
parent 50b495f3d8
commit 773aabdda6
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
server/text_generation_server/utils/gptq/custom_autotune.py
View File

@ -90,7 +90,7 @@ class Autotuner(triton.KernelInterface):
return triton.testing.do_bench( return triton.testing.do_bench(
kernel_call, percentiles=(0.5, 0.2, 0.8), rep=40 kernel_call, percentiles=(0.5, 0.2, 0.8), rep=40
) )
except triton.compiler.OutOfResources: except triton.OutOfResources:
return (float("inf"), float("inf"), float("inf")) return (float("inf"), float("inf"), float("inf"))
def run(self, *args, **kwargs): def run(self, *args, **kwargs):