A Transfer Learning Evaluation of Deep Neural Networks for Image Classification
For practitioners choosing pre-trained models for image classification, this provides a comparative evaluation but is incremental as it applies existing methods to new datasets.
This paper evaluates eleven pre-trained deep neural networks on five target datasets for image classification, measuring accuracy, training time, and model size to guide model selection. The best model achieved 97.8% accuracy on one dataset, with significant variations in performance across domains.
Transfer learning is a machine learning technique that uses previously acquired knowledge from a source domain to enhance learning in a target domain by reusing learned weights. This technique is ubiquitous because of its great advantages in achieving high performance while saving training time, memory, and effort in network design. In this paper, we investigate how to select the best pre-trained model that meets the target domain requirements for image classification tasks. In our study, we refined the output layers and general network parameters to apply the knowledge of eleven image processing models, pre-trained on ImageNet, to five different target domain datasets. We measured the accuracy, accuracy density, training time, and model size to evaluate the pre-trained models both in training sessions in one episode and with ten episodes.