Proofs for parameter calculation for isotropic kernels with uniform dilation of 1
Abstract
For a channel multiplier m, I try to show one can have an algorithm which can find parameters stride, input padding, output padding, where output size s is a multiple of input size by factor of m, alternatively one can set the output dimension to be m ×input dimension and figure out the rest of the parameters: stride, input padding and output padding correctly.
Fields of Study & Subjects
Advanced Neural Network Applications
Neural Networks and Applications
Computer Science and Engineering
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 12 February 2026
- Language
- EN