Simon Willison’s Weblog

Saturday, 26th January 2019

websocketd (via) Delightfully clever piece of design: “It’s like CGI, twenty years later, for WebSockets”. Simply run “websocketd --port=8080 my-program” and it will start up a WebSocket server on port 8080 and fire up a new process running your script every time it sees a new WebSocket connection. Standard in and standard out are automatically hooked up to the socket connection. Since it spawns a new process per connection this won’t work well with thousands of connections but for smaller scale projects it’s an excellent addition to the toolbok—and since it’s written in Go there are pre-compiled binaries available for almost everything. # 2:38 am

Practical Deep Learning for Coders 2019 (via) The deep learning evening course I took a few months ago has now been shared online in full, and it’s outstanding. “After the first lesson you’ll be able to train a state-of-the-art image classification model on your own data”—can confirm: after just the first lesson I built a bobcat v.s. cougar classifier using photos from iNaturalist. The biggest thing I learned from the course is how powerful transfer learning is. I used to think you needed a huge amount of data to get good results from deep learning. That’s no longer true: you can take an existing model (eg ResNet for image classification) and train on top of it. ResNet can classify images as 1,000 classes (house, cat, etc)—training an extra few hundred images of e.g. Bobcats vs Cougars only takes a couple of minutes on a GPU and can give you crazily accurate results. It works because the pre-trained model can already pick up really subtle details—fur patterns, ear shapes etc—so you only need to train a few more layers on it for it to be able to classify against the patterns in your new set of training images. And this doesnt just work for image classification! Natural language processing benefits from transfer learning too: take an existing model trained on the entire corpus of Wikipedia (so it knows patterns from sentence structures) and you can build IMDB sentiment analysis on top. That’s in lesson 4. # 12:32 am

2019 » January

MTWTFSS
 123456
78910111213
14151617181920
21222324252627
28293031