@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / c64612802b
Fork 0

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, 02:00 UTC

[Platform] improve platforms getattr (#12264)

Signed-off-by: Mengqing Cao <cmq0113@163.com>

Mengqing Cao committed Jan 21, 2025 at 22:42 UTC c64612802b72f1e015b98e0c569367e12391f579
1 file changed +4 -1
vllm/platforms/__init__.py
+4 -1
@@ -217,8 +217,11 @@ def __getattr__(name: str):
217 global _init_trace
218 _init_trace = "".join(traceback.format_stack())
219 return _current_platform
220 - else:
220 + elif name in globals():
221 return globals()[name]
222 + else:
223 + raise AttributeError(
224 + f"No attribute named '{name}' exists in {__name__}.")
225
226
227 __all__ = [