Offshoot of a DataCamp project to use alternative larger dataset (downloaded from Kaggle) and include entire alphabet.
While the original DC project model classifies letters within A, B, or C to 95% accuracy, this extended model classifies letters within A-Z to 89% accuracy.
Data downloaded from Kaggle, stored in Github.
Notebook written in/for Colab.