Skip to content

Latest commit

 

History

History
15 lines (9 loc) · 1.44 KB

README.md

File metadata and controls

15 lines (9 loc) · 1.44 KB

The Digit Recognition Web App - Which Digit

This is a Web Application that can recognize which numbers were drawn by the user on the canvas. For making this, it uses Deep Learning techniques to predict the resulting images. It was trained using two python frameworks for Machine Learning: Tensorflow, and Keras. To deploy the model, another framework for JavaScript was used then, the TensorflowJS. To test the web application click the link below:

Digit Recognition Web App

For more information about the model used for inference, click on the Model Card.

Main References:

  • Michael A. Nielsen, "Neural Networks and Deep Learning", Determination Press, 2015 - neuralnetworksanddeeplearning.com
  • Juan De Dios Santos Rivera, "Practical TensorFlow JS Learning Development"
  • Karen Simonyan & Andrew Zisserman, VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION, Visual Geometry Group, Department of Engineering Science, University of Oxford - VGG Paper
  • Learning Rate Reduction on Plateau - Kaggle Notebook Link