apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-09-27 21:02:24 +08:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

fix spelling error (#2827)

Browse Source
This commit is contained in:
lddfym
2025-07-14 13:12:57 +08:00
committed by GitHub
parent bad53c6b6e
commit ece88596ed
2 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/entrypoints/openai/api_server.py
View File

@@ -348,7 +348,7 @@ def reset_scheduler():
if llm_engine is None:
return Response("Engine not loaded", status_code=500)
llm_engine.scheduler.reset_scheduler()
llm_engine.scheduler.reset()
return Response("Scheduler Reset Successfully", status_code=200)
@@ -366,7 +366,7 @@ def control_scheduler(request: ControlSchedulerRequest):
return JSONResponse(content=content.model_dump(), status_code=500)
if request.reset:
llm_engine.scheduler.reset_scheduler()
llm_engine.scheduler.reset()
if request.load_shards_num or request.reallocate_shard:
if hasattr(llm_engine.scheduler, "update_config") and callable(llm_engine.scheduler.update_config):

12
fastdeploy/scheduler/config.py
View File

@@ -110,7 +110,7 @@ class GlobalSchedulerConfig:
ttl: int = 900,
min_load_score: float = 3,
max_model_len: int = 8192,
load_shrads_num: int = 1,
load_shards_num: int = 1,
enable_chunked_prefill: bool = False,
max_num_partial_prefills: int = 1,
max_long_partial_prefills: int = 1,
@@ -129,7 +129,7 @@ class GlobalSchedulerConfig:
ttl: Time-to-live in seconds for Redis keys (default 900s)
min_load_score: Minimum load score for task assignment (default 3)
max_model_len: Maximum model context length in tokens
load_shrads_num: Number of load balancing shards
load_shards_num: Number of load balancing shards
enable_chunked_prefill: Whether to enable chunked prefill processing
max_num_partial_prefills: Max partial prefill operations allowed
max_long_partial_prefills: Max long-running partial prefill ops
@@ -147,7 +147,7 @@ class GlobalSchedulerConfig:
self.topic = topic
self.ttl = ttl
self.min_load_score = min_load_score
self.load_shrads_num = load_shrads_num
self.load_shards_num = load_shards_num
self.max_model_len = max_model_len
self.enable_chunked_prefill = enable_chunked_prefill
@@ -169,8 +169,8 @@ class GlobalSchedulerConfig:
raise ValueError("ttl should be greater than 60")
if self.min_load_score < 1:
raise ValueError("min_load_score should be greater than 0")
if self.load_shrads_num < 1:
raise ValueError("load_shrads_num should be greater than 0")
if self.load_shards_num < 1:
raise ValueError("load_shards_num should be greater than 0")
r = redis.Redis(self.host, self.port, self.db, self.password)
try:
@@ -262,7 +262,7 @@ class SchedulerConfig:
topic=self.config.topic,
ttl=self.config.ttl,
min_load_score=self.config.min_load_score,
load_shrads_num=self.config.load_shrads_num,
load_shards_num=self.config.load_shards_num,
enable_chunked_prefill=self.config.enable_chunked_prefill,
max_num_partial_prefills=self.config.max_num_partial_prefills,
max_long_partial_prefills=self.config.max_long_partial_prefills,