Interactive Handwriting Recognition Model

To further explore deep learning, I made a deep learning model based on the MNIST database. The MNIST database is a set of 60,000 images of handwritten numbers from zero to nine. The model I created predicts and identifies which handwritten number is being shown. Using TensorFlow and Keras, I trained the model by implementing a convolutional neural network with various layers. Prior to training the model, I used Pandas and NumPy to apply statistical analysis to ensure the dataset is effectively organized and shuffled, and to one-hot encode the data from the model. After training, I attained a prediction accuracy of 98%. I then created a graphical user interface that allows users to draw numbers from zero to nine and receive a live prediction from the model. Overall, this project encorporates a deep learning model to a graphical user interface and allows users to interactively see how a deep learning model can work, where it can be improved, and how it can be useful.