google-deepmind / google-deepmind/dnc
Why the memory is a 3D tensor?
Open
- Dominant language
- Python
- Stars
- 2.5k
- Forks
- 446
- PR merge metrics
- No merged PRs in 30d
Description
Dear author:
We notice that in the paper, the memory is a 2D matrix (cell number, memory size]), but in pratice, the memory is a 3D tensor ([batch size, cell number, memory size]). Maybe I ignore some import detail. Could you help me address my confusion? Thank you
Contributor guide
Assessment
This issue has not been assessed yet.