25th August 2019

This post is the fourth in a series on deep learning using Keras. We've already looked at dense networks with category embeddings, convolutional networks, and recommender systems. For this installment we're going to use recurren…

29th April 2019

In this post we'll continue the series on deep learning by using the popular Keras framework to build a recommender system. This use case is much less common in deep learning literature than things like image classifiers or text…

8th January 2019

In my last post, I kicked off a series on deep learning by showing how to apply several core neural network concepts such as dense layers, embeddings, and regularization to build models using structured and/or time-series data. I…

14th October 2018

This post marks the beginning of what I hope to become a series covering practical, real-world implementations using deep learning. What sparked my motivation to do a series like this was Jeremy Howard's awesome fast.ai courses,…

2nd September 2017

Prophet is an open source forecasting tool built by Facebook. It can be used for time series modeling and forecasting trends into the future. Prophet is interesting because it's both sophisticated and quite easy to use, so it's po…