The MNIST handwritten digit classification task is the "hello world" of computer vision. Instrumental in showing the capabilities of neural networks for the field, these days it serves as a prime teaching example due to its simle data structure and relatable, useful and easily comprehensible task. A key benefit of it is that the problem gets out of the way, allowing learners to focus on the methods and their complexities, rather than having to deal with the barrage of fringe issues that quickly appears in a more complex computer vision task.

This repository is a learning repository for myself. I use it to acquaint myself with different machine learning frameworks. I chose MNIST as sample task because of its "out-of-the-way" nature, as well as its small size, making it possible to work with it on a notebook.