Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers https://arxiv.org/abs/1802.00124

No description available

This project aims to classify images from the UltraMNIST dataset, which comprises large-scale images with multiple digits per image. The task is to predict the sum of the digits present in each image, ranging from 0 to 27. The project implements Single Stage Pruned Networks (SSPNs) using BatchNorm pruning of EfficientNets.