(Pytorch) Training ResNets on ImageNet-100 data
-
Updated
Apr 1, 2026 - Python
(Pytorch) Training ResNets on ImageNet-100 data
Creates subsets of ImageNet (e.g. ImageNet100)
Systematic empirical optimization and architectural forensics of a custom ResNet-18. Features a 5-round progressive ablation protocol, from-scratch normalization implementations (BN, LN, GN, BIN), and Grad-CAM interpretability analysis on ImageNet-100.
Add a description, image, and links to the imagenet-100 topic page so that developers can more easily learn about it.
To associate your repository with the imagenet-100 topic, visit your repo's landing page and select "manage topics."