Resnet18 Model With Sequential Layer For Computing Accuracy On Image Classification Dataset

Abstract

This residual network has been a broad domain of research in deep learning. Many complex architectures are based upon residual networks. Residual networks are efficient due to skip connections. This paper highlights the addition of a sequential layer to the traditional RESNET 18 model for computing the accuracy of an Image classification dataset. The classification datasets such as Intel Scene dataset, CIFAR10 dataset, etc. These datasets consist of images belonging to various classes. In classification, we assign the pictures to their respective categories. The addition of a sequential layer gives the accuracy in the range of 0 to 1, which helps find the accuracy of prediction for the test set.

Tasks

Reproductions