Applied-Deep-Learning

This repo has a number of very useful applied deep learning notebooks. The following notebooks include:

1. columbia-landmarks - building a dnn model to classify manually collected pictures of 5 landmarks on Columbia's campu
2. color-classification - using rnn and dnn models to improve a color classification linear model
3. hyperparameter-experiments - playing with various deep learning model hyperparameters to see how performance (validation accuracy) can be improved.
4. fashion-mnist - simply mnist classification tutorial with dnn model
5. browser-classification - text classifier build to work using javascript in browser based on a pre-trained model