Evaluating ResNeXt Model Architecture for Image Classification
This is an incremental study for image classification researchers, focusing on hyper-parameter tweaks in a known model.
The paper implemented and evaluated the ResNeXt model architecture on CIFAR-10 subsets, finding that slight decreases in depth or base-width hyper-parameters did not significantly affect performance, yielding comparable results.
In recent years, deep learning methods have been successfully applied to image classification tasks. Many such deep neural networks exist today that can easily differentiate cats from dogs. One such model is the ResNeXt model that uses a homogeneous, multi-branch architecture for image classification. This paper aims at implementing and evaluating the ResNeXt model architecture on subsets of the CIFAR-10 dataset. It also tweaks the original ResNeXt hyper-parameters such as cardinality, depth and base-width and compares the performance of the modified model with the original. Analysis of the experiments performed in this paper show that a slight decrease in depth or base-width does not affect the performance of the model much leading to comparable results.