Web Environment for the Design, Implementation, Training, Evaluation and Analysis of Convolutional Neural Network Models

  • Nelson E. Vera-Parra, César A. Perdomo Ch., Julián R. Camargo L.

Abstract

This document aims to present the design, implementation and evaluation of a web tool (DeCNN - Design CNN App Web) for the design, training and evaluation of models of convolutional neural networks (CNN) using the open source library of Python: Keras. Results indicate that the tool presents a CNN design web environment with evaluation and performance analysis resources that allow for feedback that efficiently assists the design process. Additionally, DeCNN allows exporting the generated models to be used in other tools.

Published
2020-10-03
How to Cite
Nelson E. Vera-Parra, César A. Perdomo Ch., Julián R. Camargo L. (2020). Web Environment for the Design, Implementation, Training, Evaluation and Analysis of Convolutional Neural Network Models. International Journal of Advanced Science and Technology, 29(04), 10023 -. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/33041