You are here: Home » NewsFeeds » How HBO’s Silicon Valley Built “Not Hotdog” with TensorFlow, Keras and React Native

How HBO’s Silicon Valley Built “Not Hotdog” with TensorFlow, Keras and React Native

3. The DeepDog ArchitectureDesignOur final architecture was spurred in large part by the publication on April 17 of Google’s MobileNets paper, promising a new neural architecture with Inception-like accuracy on simple problems like ours, with only 4M or so parameters. This meant it sat in an interesting sweet spot between a SqueezeNet that had maybe been overly simplistic for our purposes, and the possibly overwrought elephant-trying-to-squeeze-in-a-tutu of using Inception or VGG on Mobile. The paper introduced some capacity to tune the size & complexity of network specifically to trade memory/CPU consumption against accuracy, which was very much top of mind for us at the time.With less than a month to go before the app had to launch we endeavored to reproduce the paper’s results. This was entirely anticlimactic as within a day of the paper being published a Keras implementation was already offered publicly on GitHub by Refik Can Malli,


 

Original article