ASOC 2022: Sparse metric and pattern implementation of SOFA
- Dominant language
- Python
- Stars
- 2k
- Forks
- 301
- PR merge metrics
- No merged PRs in 30d
Description
# Background
This is an advance subject of [ASoC 2022](https://opensource.alibaba.com/asoc2022) and https://github.com/alibaba/AliceMind/issues/44 .
At present, SOFA cannot support the sparse algorithm. It is necessary to refer to the previous work to implement common sparse metrics (e.g. MaP and MvP) and sparse patterns (e.g. unstructured pruning, block-wise pruning).
# Target
The implementation of common sparse metrics (e.g. MaP and MvP) and sparse patterns (e.g. unstructured pruning, block-wise pruning).
# Difficulty
Hard
# Mentor
Chuanqi Tan (@Chuanqi1992 )([chuanqi.tcq@alibaba-inc.com](mailto:chuanqi.tcq@alibaba-inc.com))
# 背景
这是一个[阿里巴巴编程之夏 2022](https://opensource.alibaba.com/asoc2022) 的基础课题 https://github.com/alibaba/AliceMind/issues/44
AliceMind-SOFA框架不支持稀疏算法,需要参考先有工作实现MaP、MvP、L0等稀疏Metric和随机稀疏、block稀疏等稀疏Pattern。
# 目标
实现MaP、MvP、L0等稀疏Metric和随机稀疏、block稀疏等稀疏Pattern。
# 难度
高
# 导师
Chuanqi Tan (@Chuanqi1992 )([chuanqi.tcq@alibaba-inc.com](mailto:chuanqi.tcq@alibaba-inc.com))
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.