GrouPy icon indicating copy to clipboard operation
GrouPy copied to clipboard

Group Equivariant Convolutional Neural Networks

Results 16 GrouPy issues
Sort by recently updated
recently updated
newest added

Hi I encounter a problem with this after setting up the environment. I wonder if you can offer an suggestion. Thank you in advance. -------------------------------------------------------------------------------- CuPy (cupy) version 6.0.0 may...

Hi Taco, Here's a tf2-compatible version of GrouPy with some very minimal changes. I understand that you're not taking PRs for GrouPy anymore, but I figure that leaving an unmerged...

Dear Taco, This PR includes the Tensorflow version in the requirements.txt. I got the code to work with TF 1.13. That might be useful to know for other users of...

Thank you for giving a fascinating tutorial in NeurIPS 2020 about equivariance, I am very interested and new in this field. After reading your paper, I have a question about...

Hello Dr.Cohen, this is another question. I didn't catch the meaning of GArray and GFuncArray: A GFuncArray is an array of functions on a group G. A GArray represents an...

I'm puzzled by the number of trainable parameters in networks using `gconv2d`. The script below creates a network using `gconv2d`s from Z2 to C4 to C4 and counts the number...

Hey Taco, The paper [3D G-CNNs for Pulmonary Nodule Detection](https://arxiv.org/pdf/1804.04656.pdf) refers to this repository for an opensource implementation of the `GConv3D` function. I found the 2D variant, but no 3D...

I implemented group equivariant convolutions in PyTorch. Code includes tests for filter transformation and equivariance. I also tested this implementation by training on CIFAR-10 with ResNet-18 architecture https://github.com/adambielski/pytorch-gconv-experiments

This adds P4-conv and P4M-conv classes for PyTorch. The implementation is based on the Chainer implementation (SplitGConv2D base class), the main difference being that the filter transforms are implemented using...

The following changes allowed me to run gconv_experiments using: - chainer (5.1.0) - cupy-cuda90 (5.1.0) - Python 3.6.7 Addresses many of the issues mentioned in #13 . Feel free to...