Graphical User Interface for Tensorflow
- Mentors
- Monjoy Saha, Pooya Mobadersany
- Organization
- Biomedical Informatics, Emory University
The idea of this project is to make a software in which a user can make deep learning models in an easy way using a graphical user interface with backend supported by tensorflow. Through the graphical user interface, a user will able to add, delete, edit deep learning layers in a model. The main purpose of the project is to make the implementation of deep learning models quick and easy.
The software will be built using electron-js which is a framework for building cross-platform desktop apps with HTML, CSS, and JavaScript. It will have a drag-and-drop feature to build deep learning models in the form of a graph which will then converted to a python code by the software. The generated code will then be executed in the child process which trains the deep learning model and sends the metrics data( loss, accuracy ) to the parent process which then plots the statistics.