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, 06:45 UTC
Fix type hints (#1427)
Light Lin committed
Oct 20, 2023 at 23:50 UTC
f61dc8072fbbef76430291ef62d79b7dd265b0e3
1 file changed
+1
-1
vllm/core/scheduler.py
+1
-1
@@ -275,7 +275,7 @@ class Scheduler:
275
# Create input data structures.
276
seq_group_metadata_list: List[SequenceGroupMetadata] = []
277
for seq_group in scheduler_outputs.scheduled_seq_groups:
278
- seq_data: Dict[int, List[SequenceData]] = {}
278
+ seq_data: Dict[int, SequenceData] = {}
279
block_tables: Dict[int, List[int]] = {}
280
for seq in seq_group.get_seqs(status=SequenceStatus.RUNNING):
281
seq_id = seq.seq_id