An experimental comparison of the widely used pre‐trained deep neural networks for image classification tasks towards revealing the promise of transfer‐learning

• Published in: Concurrency and computation Vol. 34; no. 24
• Main Authors: Kabakus, Abdullah Talha, Erdogmus, Pakize
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.11.2022; Wiley Subscription Services, Inc
• Subjects: Artificial neural networks; Benchmarks; convolutional neural network; Deep learning; deep neural network; Image classification; Keras; Machine learning; Neural networks; TensorFlow; transfer‐learning
• DOI: 10.1002/cpe.7216
• ISSN: 1532-0626

Summary The easiest way to propose a solution based on deep neural networks is using the pre‐trained models through the transfer‐learning technique. Deep learning platforms provide various pre‐trained deep neural networks that can be easily applied for image classification tasks. So, “Which pre‐trained model provides the best performance for image classification tasks?” is a question that instinctively comes to mind and should be shed light on by the research community. To this end, we propose an experimental comparison of the six popular pre‐trained deep neural networks, namely, (i) VGG19, (ii) ResNet50, (iii) DenseNet201, (iv) MobileNetV2, (v) InceptionV3, and (vi) Xception by employing them through the transfer‐learning technique. Then, the proposed benchmark models were both trained and evaluated under the same configurations on two gold‐standard datasets, namely, (i) CIFAR‐10 and (ii) Stanford Dogs to benchmark them. Three evaluation metrics were employed to measure performance differences between the employed pre‐trained models as follows: (i) Accuracy, (ii) training duration, and (iii) inference time. The key findings that were obtained through the conducted a wide variety of experiments were discussed.