[Feature] Control the stage init timeout threshold by --stage-init-timeout (#393)
Signed-off-by: Taichang Zhou <tzhouam@connect.ust.hk> Signed-off-by: Zhou Taichang <tzhouam@connect.ust.hk> Signed-off-by: tzhouam <tzhouam@connect.ust.hk> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: yinpeiqi <yinpeiqi809@gmail.com> Signed-off-by: Li Shunyang <gmsv2544@nus.edu.sg> Signed-off-by: Li Shunyang <lishunyang12@163.com> Signed-off-by: Peiqi Yin <60515999+yinpeiqi@users.noreply.github.com> Signed-off-by: dengyunyang <584797741@qq.com> Signed-off-by: Samit <285365963@qq.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Canlin Guo <canlinguosdu@gmail.com> Co-authored-by: Peiqi Yin <60515999+yinpeiqi@users.noreply.github.com> Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com> Co-authored-by: SYLAR <125541396+lishunyang12@users.noreply.github.com> Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com> Co-authored-by: dengyunyang <584797741@qq.com> Co-authored-by: Samit <285365963@qq.com>
This commit is contained in:
11 files changed
+64
-58
No files matched your search
@@ -322,7 +322,7 @@ def main(args):
|
||||
omni_llm = Omni(
|
||||
model=model_name,
|
||||
log_stats=args.enable_stats,
|
||||
init_sleep_seconds=args.init_sleep_seconds,
|
||||
stage_init_timeout=args.stage_init_timeout,
|
||||
batch_timeout=args.batch_timeout,
|
||||
init_timeout=args.init_timeout,
|
||||
shm_threshold_bytes=args.shm_threshold_bytes,
|
||||
@@ -426,10 +426,10 @@ def parse_args():
|
||||
help="Enable writing detailed statistics (default: disabled)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--init-sleep-seconds",
|
||||
"--stage-init-timeout",
|
||||
type=int,
|
||||
default=20,
|
||||
help="Sleep seconds after starting each stage process to allow initialization (default: 20)",
|
||||
default=300,
|
||||
help="Timeout for initializing a single stage in seconds (default: 300)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--batch-timeout",
|
||||
|
||||
@@ -224,6 +224,7 @@ def main(args):
|
||||
model=model_name,
|
||||
stage_configs_path=args.stage_configs_path,
|
||||
log_stats=args.enable_stats,
|
||||
stage_init_timeout=args.stage_init_timeout,
|
||||
)
|
||||
|
||||
thinker_sampling_params = SamplingParams(
|
||||
@@ -337,10 +338,10 @@ def parse_args():
|
||||
help="Enable writing detailed statistics (default: disabled)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--init-sleep-seconds",
|
||||
"--stage-init-timeout",
|
||||
type=int,
|
||||
default=20,
|
||||
help="Sleep seconds after starting each stage process to allow initialization (default: 20)",
|
||||
default=300,
|
||||
help="Timeout for initializing a single stage in seconds (default: 300)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--batch-timeout",
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
python end2end.py --output-wav output_audio \
|
||||
--query-type use_audio \
|
||||
--init-sleep-seconds 90
|
||||
--stage-init-timeout 300
|
||||
|
||||
# init-sleep-seconds works to avoid two vLLM stages initialized at the same time within a card.
|
||||
# stage-init-timeout sets the maximum wait to avoid two vLLM stages initializing at the same time on the same card.
|
||||
Reference in new issue
Block a user