huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-12 04:44:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix GPTQ autotune data type to be compatible with Torch 2.4.0

Browse Source
This commit is contained in:
Daniël de Kok 2024-07-25 09:39:42 +00:00
parent 26614057a7
commit fa9221f28d
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/layers/gptq/custom_autotune.py
View File

@ -91,7 +91,7 @@ class Autotuner(triton.KernelInterface):
kernel_call, quantiles=(0.5, 0.2, 0.8), rep=40 kernel_call, quantiles=(0.5, 0.2, 0.8), rep=40
) )
except triton.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):
self.nargs = dict(zip(self.arg_names, args)) self.nargs = dict(zip(self.arg_names, args))