- Created Adding ActivityPub to your static site and also added a lot more functionality to my site. I can now show Likes, "Boosts" and replies.
- See Interactions from around the fediverse with https://paul.kinlan.me/adding-activity-pub-to-your-static-site/ as an example
- Added resources to ActivityPub to help people find the best information
- ML
- Google Colab - Colab is a rather amazing tool, I've also built some dashboards in the past with it but for ML with Tensorflow it's worked really well.
- I've started my first project to help me learn GitHub - PaulKinlan/button-and-link-scraper the goal is to check to see an anchor on a page looks like a button, and should possibly be a button. It's a small A11Y feature, but something that I think is hard to test just from the HTML.
- I had a lot of fun building this even before I got to training any NN, for a list of URLs I find all the button-like elements and screenshot them, and I also find all the links too and screenshot them.
- The NN is just using a simple image classifier, right now for anything that looks like a button it has a 100% accuracy, which I am sure is suspicious.
- Lots of research
- AI and Machine Learning For Coders: A Programmer's Guide to Artificial Intelligence: Amazon.co.uk: 9781492078197: Books by Laurence Moroney - I'm familiar with the basics of NN and this book got me back up to speed and the latest developments (Convolution Networks, Recurrent Networks etc). I was really good to get a sense about how image classification works, Image Segmentation, Text prediction- I had no clue but a lot of it is quite logical on retrospect.
- This talk from IO complements the book - Machine Learning Zero to Hero (Google I/O'19) - YouTube
- TensorFlow
- Transformers - GPT, BERT, Copilot and others use this and it seems like a promising area that I should at least know the basics of.
- Started to do some research because I'd heard about it - this video was a good intro [Transfer learning and Transformer models (ML Tech Talks) - YouTube](Transfer learning and Transformer models (ML Tech Talks) - YouTube) and this book is great too https://www.amazon.co.uk/gp/product/1098103246/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1 - both give a good overview, personally I found https://arxiv.org/abs/1706.03762 a bit over my head.
- Other bits:
- Grokking Deep Reinforcement Learning: Amazon.co.uk: Morales, Miguel: 9781617295454: Books
- Jay Alammar – Visualizing machine learning one concept at a time. < Has a good post about how Stable Diffusion works.
- AI and Machine Learning For Coders: A Programmer's Guide to Artificial Intelligence: Amazon.co.uk: 9781492078197: Books by Laurence Moroney - I'm familiar with the basics of NN and this book got me back up to speed and the latest developments (Convolution Networks, Recurrent Networks etc). I was really good to get a sense about how image classification works, Image Segmentation, Text prediction- I had no clue but a lot of it is quite logical on retrospect.
I lead the Chrome Developer Relations team at Google.
We want people to have the best experience possible on the web without having to install a native app or produce content in a walled garden.
Our team tries to make it easier for developers to build on the web by supporting every Chrome release, creating great content to support developers on web.dev, contributing to MDN, helping to improve browser compatibility, and some of the best developer tools like Lighthouse, Workbox, Squoosh to name just a few.
