Machine Learning is Fun! Part 4: Modern Face Recognition with Deep Learning
Adam Geitgey
5K107

How do you build a scalable classifier? I’m thinking of something like AWS Rekognition, which can build up a database of embeddings which you can then query with an embedding and it will give you all matches with > 50%. Do they continually train an SVM? Or do they use euclidian distance brute force?

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.