Randall Hunt is a Senior Technical Evangelist and Software Engineer at Amazon Web Services in Los Angeles. Randall spends most of his time building demos and writing about new services and launches on the AWS News Blog. Python is his favorite programming language but he can sometimes be found in the dark realm of C++. Prior to working at AWS, Randall launched rockets at NASA and SpaceX but he found his programming passion at MongoDB. He is a total space nerd.
Learn how we designed, built, and deployed the @WhereML Twitter bot that can identify where in the world a picture was taken using only the pixels in the image. We'll dive deep on artificial intelligence and deep learning with the MXNet framework and also talk about working with the Twitter Account Activity API.
The WhereML twitter bot is built on the LocationNet model which is trained with the Berkley Multimedia Commons public dataset of 33.9 million geotagged images from Flickr. The model is based on a ResNet-101 architecture and adds a classification layer that splits the earth into ~15000 cells created with Google's S2 spherical geometry library. This model is based on prior work at Berkley and Google.
In this session we'll start by describing AI in general terms then diving into deep learning and the MXNet framework. We'll describe the LocationNet model in detail and show how it is trained and created in Amazon SageMaker. Finally we'll talk about the Twitter Account Activity webhooks API and how to interact with it using an API Gateway and AWS Lambda function.
Attendees are encouraged to interact with the bot in real-time at whereml.bot or on twitter at @WhereML
All code used in this project is open source and attendees are encouraged to experiment with it.