flâneur

Scale-invariant-CNNs/Implementation.md at master · wsgdrfz/Scale-invariant-CNNs · GitHub

github.com · 957 words · saved by 1 readers

Study the Scale Invariance or Equivariance Convolutional Neural Network - wsgdrfz/Scale-invariant-CNNs

SIE-CNNs Implementation (PyTorch) All the implementation can be found in our GitHub. All the coding is done in PyTorch, and the code introduced in this documentation can be found under src/pytorch/. Code Sturcture Make_datasets_Scale.py: Generate scale dataset from the original one, for example, from MNIST to scale-MNIST. Standard_CNN.py: The baseline, a standard convolutional neural network with 3 convolution layers and 2 fully connected layers. SI_ConvNet.py: The implementation of [1]. class ScaleInvariance_Layer: The scale-invariance layer inherit from nn.Module including the…

saved by

related reading