11.2. Attention Pooling by Similarity — Dive into Deep Learning 1.0.3 documentation
d2l.ai
one-hot-encoding if we shift and rotate ifferent notions of range and smoothnes Gaussian kernel with one of a different width according to how many similar observations are available
2 Users
0 Comments
15 Highlights
0 Notes
Top Highlights
- one-hot-encoding
- if we shift and rotate
- ifferent notions of range and smoothnes
- Gaussian kernel with one of a different width
- according to how many similar observations are available
Domain
Ready to highlight and find good content?
Glasp is a social web highlighter that people can highlight and organize quotes and thoughts from the web, and access other like-minded people’s learning.