cnn-explainer

Why we included this project

Most explanations of convolutional networks are either dense math or hand-waving diagrams. CNN Explainer is neither: it runs a real pretrained tiny-VGG model in the browser and lets you click through the layers to watch kernels, feature maps, and activations respond to live input images. It was built for people who are not deep-learning experts, so the visualizations carry the explanation instead of equations. That makes it a useful classroom demo for instructors and a low-stakes way for developers to build intuition before they touch PyTorch or TensorFlow. The code and trained model are both included, so you can run it locally or adapt it to your own model and image classes.