apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-06 00:57:33 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Feature] Support repetition early stop (#3024)

Browse Source 
* support repetition early stop and support user to set the parameter

* remove log

* fix codestyle

* add the early_stop_config to rollout_config

* update config and EarlyStopper class

* fix the bug for triton

* modify the stop method

* update description

* modify the usage for stop_flags

---------

Co-authored-by: Yuanle Liu <yuanlehome@163.com>
This commit is contained in:
Zero Rains
2025-07-29 22:42:54 +08:00
committed by GitHub
parent 3214fb5393
commit b2f9a42d87
13 changed files with 575 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/model_executor/layers/sample/meta_data.py
View File

@@ -44,5 +44,7 @@ class SamplingMetadata:
top_k: Optional[paddle.Tensor] = None
min_p: Optional[paddle.Tensor] = None
max_num_logprobs: Optional[int] = None
enable_early_stop: Optional[int] = False
stop_flags: Optional[paddle.Tensor] = None
prompt_ids: Optional[paddle.Tensor] = None
prompt_lens: Optional[paddle.Tensor] = None