Custom PyTorch Conv1d/Conv2d layers with global context conditioning, group-wise operations, and depthwise convolution support.
-
Updated
May 13, 2025 - Python
Custom PyTorch Conv1d/Conv2d layers with global context conditioning, group-wise operations, and depthwise convolution support.
ResNet vs ResNeXt on Oxford-IIIT Pet — both implemented from scratch in PyTorch, parameter-matched (11.2M vs 10.6M), trained from random init under identical conditions. Includes misclassification overlap analysis and hardware benchmarks.
Add a description, image, and links to the grouped-convolution topic page so that developers can more easily learn about it.
To associate your repository with the grouped-convolution topic, visit your repo's landing page and select "manage topics."