In this project we built an email classifier using neural networks. The purposes of the project are educative. The project focuses on fine-tuning hyperparameters for a base neural network and reduce variance on the development set.
We explain (not in depth) concepts like Normalization, Batch Normalization, Dropout, learning rate and kernel initializer.
Additionally to pandas, numpy, matplotlib we used scikit-learn, tensorflow and keras-tuner.