Visualized Convolutional Neural Networks with MNIST, 2023
Seonghyeon Kim, Changwook Seo
We are excited to introduce our AI artwork titled 'Visualized Convolutional Neural Networks with MNIST.' Utilizing layers from a convolutional neural network (CNN), we have developed a classifier model that accurately classifies images from the MNIST dataset. The model's weights are parsed in Python and visualized through a processing IDE, offering a clear depiction of the convolution process. Each layer of the CNN displays the calculated values in matrix-shaped blocks, enhancing the understanding of how the model processes data.
