apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

[BugFix][Optimization] Replace silent failures with catchable exceptions and informative error messages (#6533)

Browse Source 
* init

* init

* fix format

* add

* add files

* add ut

* fix some

* add ut

* add more

* add

* fix pre-commit

* fix pre-commit

* fix cover

* skip long seq

* add

* add

* fix

* remove not need

* fix set attr

* fix comments

* fix comments

* fix failed tests

---------

Co-authored-by: gongweibao <gognweibao@baidu.com>
This commit is contained in:
gongweibao
2026-03-16 21:32:43 +08:00
committed by GitHub
parent d113397b09
commit a6351dea0b
61 changed files with 1595 additions and 171 deletions
Download Patch File Download Diff File Expand all files Collapse all files
fastdeploy/worker/hpu_model_runner.py
+3 -1
View File
@@ -1160,7 +1160,9 @@ class HPUModelRunner(ModelRunnerBase):
"""
Initialize attention backends and forward metadata
"""
assert len(self.attn_backends) == 0
assert (
len(self.attn_backends) == 0
), f"attn_backends should be empty before initialization, got {len(self.attn_backends)} backends"
# TODO(gongshaotian): Get rank from config
num_heads = self.model_config.num_attention_heads // self.parallel_config.tensor_parallel_size