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 14, 05:30 UTC
[Minor] Change warning->warning_once in preprocess (#26455)
Signed-off-by: Zhuohan Li <zhuohan123@gmail.com>
Zhuohan Li committed
Oct 8, 2025 at 21:09 UTC
ebf6ef1a9b95d3a46b952b7933c0b41061df380a
1 file changed
+2
-2
vllm/inputs/preprocess.py
+2
-2
@@ -69,7 +69,7 @@ class InputPreprocessor:
69
70
def get_bos_token_id(self) -> Optional[int]:
71
if self.tokenizer is None:
72
- logger.warning(
72
+ logger.warning_once(
73
"Using None for BOS token id because tokenizer is not initialized"
74
)
75
return None
@@ -78,7 +78,7 @@ class InputPreprocessor:
78
79
def get_eos_token_id(self) -> Optional[int]:
80
if self.tokenizer is None:
81
- logger.warning(
81
+ logger.warning_once(
82
"Using None for EOS token id because tokenizer is not initialized"
83
)
84
return None