EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[Fixbug] Fix shape not match when sliding_window and dynamic batch_size (#2830)

Browse Source 
### What this PR does / why we need it?
Fix shape not match when test LLM-Research/Phi-4-mini-instruct accuarcy 

### Does this PR introduce _any_ user-facing change?

Users can't set dynamic batch_size or use lm_eval test accuracy when
using models(sliding_window)

### How was this patch tested?
accuarcy of LLM-Research/Phi-4-mini-instruct is ok :
```
vllm (pretrained=LLM-Research/Phi-4-mini-instruct,max_model_len=4096,dtype=auto,tensor_parallel_size=1), gen_kwargs: (None), limit: None, num_fewshot: 5, batch_size: auto
|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.8105|±  |0.0108|
|     |       |strict-match    |     5|exact_match|↑  |0.8097|±  |0.0108|
```


- vLLM version: v0.10.2
- vLLM main:
3c96e7b8a1

Signed-off-by: hfadzxy <starmoon_zhang@163.com>
This commit is contained in:
zhangxinyuehfad
2025-09-19 22:35:14 +08:00
committed by GitHub
parent cf549b976d
commit a22b532d38
2 changed files with 37 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
vllm_ascend/attention/attention_v1.py
View File

@@ -378,7 +378,8 @@ class AscendAttentionBackendImpl(AttentionImpl):
# seq_lens_tensor needs to be transferred to the device for 310P.
attn_metadata.seq_lens = \
attn_metadata.seq_lens.to(device=query.device)
if self.sliding_window is not None:
if self.sliding_window is not None and attn_metadata.seq_lens.shape[
0] == query.size(0):
batch_size = attn_metadata.seq_lens.shape[0]
block_size = 128
query = query.view(batch_size, 1, self.num_heads * self.head_size)