A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 08:00 UTC
Fix typo in sampling_params.py (#788)
wangcx18 committed
Aug 18, 2023 at 09:12 UTC
0c04ce323448af8cb576f9698f907ef2e6a8c1d0
1 file changed
+2
-2
vllm/sampling_params.py
+2
-2
@@ -77,7 +77,7 @@ class SamplingParams:
77
78
self._verify_args()
79
if self.use_beam_search:
80
- self._verity_beam_search()
80
+ self._verify_beam_search()
81
elif self.temperature < _SAMPLING_EPS:
82
# Zero temperature means greedy sampling.
83
self._verify_greedy_sampling()
@@ -109,7 +109,7 @@ class SamplingParams:
109
raise ValueError(
110
f"logprobs must be non-negative, got {self.logprobs}.")
111
112
- def _verity_beam_search(self) -> None:
112
+ def _verify_beam_search(self) -> None:
113
if self.best_of == 1:
114
raise ValueError("best_of must be greater than 1 when using beam "
115
f"search. Got {self.best_of}.")