@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 0961f5a49a
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 17, 09:15 UTC

Add find method to sequence group

Woosuk Kwon committed Feb 13, 2023 at 02:39 UTC 0961f5a49a7f826a8b8be08451b308b374868c51
1 file changed +6
cacheflow/sequence.py
+6
@@ -76,3 +76,9 @@ class SequenceGroup:
76 return len(self.seqs)
77 else:
78 return len([seq for seq in self.seqs if seq.status == status])
79 +
80 + def find(self, seq_id: int) -> Sequence:
81 + for seq in self.seqs:
82 + if seq.seq_id == seq_id:
83 + return seq
84 + raise ValueError(f'Sequence {seq_id} not found.')