Sometimes you just can't tell if your hot dog is the real deal. It happens to the best of us. You sit down to enjoy your delicious hot dog but suspect it's really a hot dog in disguise.
Let those worries rest because now you can be certain of your hot dogs' authenticity. No more sleepless nights questioning what you ate for dinner. With the brand new AI-powered 'Hotdog' app you can be confident that when you order a hot dog, you're getting a hot dog!
This is a project I did for a course on neural networks. It's based on the 'Seefood' app from the HBO show Silicon Valley which identifies food as either 'hotdog' or 'not hotdog'. HBO created a real iOS and Android app based on the show. Details of the app development process can be found here.
The code here is for my take on the idea, a Shiny app. I used Keras with Tensorflow to build a model using data from Kaggle. The model development files can be found here.
Note that if you want to rebuild this app yourself you will need to train a model and place it in the same directory as the app. The model I used is not saved in this repository because the file size is too big.