Memory-efficient deep learning inference with incremental weight loading and data layout reorganization on edge systems.
Saved in:
| Title: | Memory-efficient deep learning inference with incremental weight loading and data layout reorganization on edge systems. |
|---|---|
| Authors: | Ji, Cheng1, Wu, Fan2, Zhu, Zongwei1,2, zzw1988@ustc.edu.cn, Chang, Li-Pin3, Liu, Huanghe2, Zhai, Wenjie2 |
| Source: | Journal of Systems Architecture; Sep2021, Vol. 118, pN.PAG-N.PAG, 1p |
| Database: | Applied Science & Technology Source |
| ISSN: | 13837621 |
|---|---|
| DOI: | 10.1016/j.sysarc.2021.102183 |