@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 9e05252b46
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, 07:00 UTC

[Misc] Add __setitem__ for LazyDict (#10469)

Signed-off-by: Yanyi Liu <wolfsonliu@163.com>

Yanyi Liu committed Nov 20, 2024 at 12:44 UTC 9e05252b46a92a5d14e4e6fd02b75383c5cf243b
1 file changed +3
vllm/utils.py
+3
@@ -1491,6 +1491,9 @@ class LazyDict(Mapping, Generic[T]):
1491 self._dict[key] = self._factory[key]()
1492 return self._dict[key]
1493
1494 + def __setitem__(self, key: str, value: Callable[[], T]):
1495 + self._factory[key] = value
1496 +
1497 def __iter__(self):
1498 return iter(self._factory)
1499