apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-02 07:12:31 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[stop sequence] support stop sequence (#3025)

Browse Source 
* stop seqs in multi-ends

* unittest for gpu stop op

* kernel tid==0
This commit is contained in:
JYChen
2025-07-29 14:17:37 +08:00
committed by GitHub
parent 1a815b7a2a
commit dafe02a7b9
11 changed files with 193 additions and 189 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
custom_ops/setup_ops.py
View File

@@ -260,7 +260,6 @@ elif paddle.is_compiled_with_cuda():
"gpu_ops/token_penalty_only_once.cu",
"gpu_ops/stop_generation.cu",
"gpu_ops/stop_generation_multi_ends.cu",
"gpu_ops/stop_generation_multi_stop_seqs.cu",
"gpu_ops/set_flags.cu",
"gpu_ops/update_inputs_v1.cu",
"gpu_ops/recover_decode_task.cu",
@@ -529,7 +528,6 @@ elif paddle.is_compiled_with_custom_device("iluvatar_gpu"):
sources=[
"gpu_ops/get_padding_offset.cu",
"gpu_ops/set_value_by_flags.cu",
"gpu_ops/stop_generation_multi_stop_seqs.cu",
"gpu_ops/rebuild_padding.cu",
"gpu_ops/update_inputs.cu",
"gpu_ops/stop_generation_multi_ends.cu",