The KNN digit classifier project involves classifying a subset of the MNIST handwritten digit dataset (1,2,3) with a ground-up implementation of the KNN algorithm in Python. For a proximity measurement, the Euclidean distance was used.
The final performance of my KNN classifier attained a 96% classification accuracy with k=3 neighbors and 97% with k=1 neighbor. In the figure above, it is evident that the poorly written digits had a propensity for being incorrectly classified while the better written digits were classified correctly.
Software: Python
Conceptual: K Nearest Neighbors | Supervised learning