Salt Lake City, Utah
June 23, 2018
June 23, 2018
July 27, 2018
Deep learning, a new era of machine learning technique, shows the ability to generalize computational model through hierarchical layers by learning the features from a large amount of training data without any human intervention. Deep learning was mainly implemented in areas of computer vision, audio, and speech processing. Especially deep learning has greatly improved state of art in image classification. Image-based classification, the most classic but significant research topic, has been critically improved using deep learning architecture and outperformed the majority of state-of-the-art achievements. No bells and whistles, deep learning becomes the most popular research area. However, in order to train a comprehensive classifier, a large amount of data is inescapable, which apparently needs lots of computational time so that it requires fast GPUs to accomplish fast training. In general, powerful GPUs are expensive and the process of solving compatibility problems in both hardware and software are reluctant as a consequence, universities only have very limited access to GPUs’ implementations. The main purpose of this paper is in identifying how to propagate principles of deep learning to students and build the bridge to make them learn and use deep learning easier or more accessible. In this paper, we develop a Graphic User Interface (GUI) deep learning for beginners to test classification results through CPU without massive training on computation. We use the high-performance GPU to train a model, once finished, we save the parameters of training model and load it to the GUI and provide a platform for students to do the testing.
Akundi, A., & Tseng, T. B., & Cao, Z., & Kim, H. (2018, June), A Deep Learning Graphical User Interface Application on MATLAB Paper presented at 2018 ASEE Annual Conference & Exposition , Salt Lake City, Utah. https://peer.asee.org/29674
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2018 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015