Winograd/Cook/Toom Convolution
Open
- Dominant language
- Julia
- Stars
- 258
- Forks
- 138
- Avg merge
- 2d 1h
- Merged PRs (30d)
- 3
Description
If you don't mind I am opening an issue just to track the possibility of having an implementation of the winograd convolution method that is faster than the traditional im2col + gemm method.
See https://arxiv.org/abs/1509.09308.
Also from one of the authors of that paper:
https://github.com/Maratyszcza/NNPACK/issues/8#issue-143884473
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.