squall/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

[Misc] Fix typos in scheduler.py (#7285)

Browse Source 
Signed-off-by: Rui Qiao <ruisearch42@gmail.com>
This commit is contained in:
Rui Qiao 2024-08-07 17:06:01 -07:00 committed by GitHub
parent e53dfd3eaf
commit 746709642c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/core/scheduler.py
View File

@ -678,7 +678,7 @@ class Scheduler:
all tokens. all tokens.
Returns: Returns:
SchedulerSwappedInOutputs. SchedulerPrefillOutputs.
""" """
ignored_seq_groups: List[SequenceGroup] = [] ignored_seq_groups: List[SequenceGroup] = []
seq_groups: List[SequenceGroup] = [] seq_groups: List[SequenceGroup] = []
@ -851,7 +851,7 @@ class Scheduler:
preempted=preempted, preempted=preempted,
) )
def _schedule_chunked_prefill(self): def _schedule_chunked_prefill(self) -> SchedulerOutputs:
"""Schedule queued requests. """Schedule queued requests.
Chunked prefill allows to chunk prefill requests, batch them together Chunked prefill allows to chunk prefill requests, batch them together
@ -862,7 +862,7 @@ class Scheduler:
The policy can sustain the high GPU utilization because it can put The policy can sustain the high GPU utilization because it can put
prefill and decodes requests to the same batch, while it improves prefill and decodes requests to the same batch, while it improves
inter token latency because decodes requests don't need to blocked inter token latency because decodes requests don't need to be blocked
by prefill requests. by prefill requests.
""" """
budget = SchedulingBudget( budget = SchedulingBudget(