Glossary term
Glossary term
Architecture
In a convolutional operation or pooling, the delta in each dimension of the next series of input slices. For example, the following animation demonstrates a (1,1) stride during a convolutional operation. Therefore, the next input slice starts one position to the right of the previous input slice. When the operation reaches the right edge, the next slice is all the way over to the left but one position down.
The preceding example demonstrates a two-dimensional stride. If the input matrix is three-dimensional, the stride would also be three-dimensional.
Created for this library
A computer vision team uses stride 2 in some convolutions to downsample feature maps while keeping receptive fields growing.
An autonomous-driving team tunes stride in its detector backbone to balance receptive field size against computational cost.
A medical imaging team uses small strides in its segmentation model to preserve fine spatial details in scans.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License