A Study on the feed-forward loop in ResNet network using 7-segment LED images.
This project is a self-study course for the beginners of machine learning and Python. It is based on the "Dive into Deep Learning" (https://d2l.ai/) and program is based on NumPy/MXNet. A special sets of images from so-called 7-segment LED are added to give more intuition on how the neural network works. Also, a feed-forward loop is added (equivalently, setting a special initial conditiomns on the residual network of the ResNet network.
All the basic codes are highly leveraged from the Numpy/MXNet from "Dive into Deep Learning" but some clumsy codes are added by the author as well. You are very welcome to contribute to this project to refine the codes and to add more ideas on the feed-forward loop.
Codes and related files will be uploaded soon.
Thanks...
Nicholas Kim