Recognizing Handwritten Letters with Neural Networks
The goal of this project was to create a convolutional neural network (CNN) model that can recognize handwriting from the first five letters in the alphabet (A,B,C,D,E). We had to use transfer learning in order to create a model that can do that. Transfer learning is when a machine learning model that is trained to solve one problem is applied to solve a different but related problem. In this scenario we used a CNN that was trained to recognize handwritten digits and applied it to recognizing handwritten letters. The data for the handwritten letters were obtained by friends and family.
“Example of Handwritten Letter Training Data”
“Loss and Accuracy Charts of CNN Model”