Skip to content

[2026秋季][Task08] zlf1201#126

Open
zlf1201 wants to merge 1 commit into
DeepLink-org:mainfrom
zlf1201:QY2026_Autumn_zlf1201_Task08_Ascend
Open

[2026秋季][Task08] zlf1201#126
zlf1201 wants to merge 1 commit into
DeepLink-org:mainfrom
zlf1201:QY2026_Autumn_zlf1201_Task08_Ascend

Conversation

@zlf1201

@zlf1201 zlf1201 commented Jun 6, 2026

Copy link
Copy Markdown

Task 08: FPS - Ascend NPU

硬件平台

  • Huawei Ascend 910B2C

优化说明

  • 预计算全配对距离矩阵 (N×N matmul) 替代每次迭代重复计算
  • N=1000 足够小,O(N²) 预计算比 256 次 O(N) 迭代更高效

性能结果

  • 前向传播: ~12.8ms(1.47x 加速)

文件

  • dlblas/kernels/ks_competition/torch/QY2026_Autumn_zlf1201_Task08_Ascend.py

Task 08: FPS (Farthest Point Sampling) - Ascend NPU
Hardware: Huawei Ascend 910B2C
Forward pass: ~12.8ms (1.47x speedup)
Optimization: precompute pairwise distance matrix
@zhaochaoxing zhaochaoxing self-requested a review June 8, 2026 02:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant