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 16, 19:30 UTC
[BUGFIX]: top_k is expected to be an integer. (#7227)
Atilla Akkuş committed
Aug 7, 2024 at 10:32 UTC
7b261092de3b008f7a2c218e338f4f8a025c93ee
1 file changed
+3
vllm/sampling_params.py
+3
@@ -224,6 +224,9 @@ class SamplingParams:
224
if self.top_k < -1 or self.top_k == 0:
225
raise ValueError(f"top_k must be -1 (disable), or at least 1, "
226
f"got {self.top_k}.")
227
+ if not isinstance(self.top_k, int):
228
+ raise TypeError(
229
+ f"top_k must be an integer, got {type(self.top_k).__name__}")
230
if not 0.0 <= self.min_p <= 1.0:
231
raise ValueError("min_p must be in [0, 1], got "
232
f"{self.min_p}.")