We’ve released pre-trained BigBiGAN representation learning models arxiv.org/abs/1907.02544 on TF Hub: tfhub.dev/s?publisher=de… Try them out in a Colab at: colab.research.google.com/github/tensorf…

Mathematics for Machine Learning. Looks like a nice companion book for machine learning and deep learning textbooks. They have a free PDF version on their website, which is nice :)

We're developing alternative ways to train our AI systems so that we can do more with less labeled training data overall. Learn how our “semi-weak supervision” method is delivering state-of-the-art performance for highly efficient, production-ready models. ai.facebook.com/blog/billion-s…

Spent some time playing with an Arduino tonight, and can confirm both examples from blog post (speech recognition, and gesture classification) work as expected. Very cool!

I'm really happy w/ caching in tf.data. Have expensive preprocessing in your pipeline? ds = tf.​data.Dataset.list_files('images/*') ds = ds.​map(load_and_preprocess) A cache lets you pay for that just once. The speedup is dramatic, and takes one LOC. Example:

We are happy to announce the v2.0 release of the Google Research Football Environment. The most exciting feature of this release is the Game Server, which lets your agent compete online with other researchers' models. Visit research-football.dev and give it a try!

Google Colab has now a Subscription model for Power users: - Faster GPUs - Longer runtimes - More memory It's for $9.99/Month. I know many power users might enjoy it: colab.research.google.com/signup

The first release candidate for UMAP 0.4 is out providing lots of new features, including performance improvements, embedding to different manifolds, inverse transform, and plotting tools.

Great one from two of my favourites @OlivierSeynnes and Neil Cronin Simple muscle architecture (SMA) analysis. I’ll definitely use this too from Monday. journals.plos.org/plosone/articl…

If you’re interested in learning about Machine Learning on Microcontrollers, check out this helpful TinyML book by Pete Warden and Daniel Situnayake! 🐦 Learn more ↓ goo.gle/2V3h1O8

A key design principle I follow in libraries (e.g. Keras) is "progressive disclosure of complexity". Make it easy to get started, yet make it possible to handle arbitrarily flexible use cases, only requiring incremental learning at each step. Like zooming in a complex landscape.

#raschka

I love it 😀

Four Strategies for Your Emotion Management Playbook Learn more 👉 Playbook.mitsmr.com/2MAYoPr

That's a great release! - AMD GPU support via ROCm (binaries available directly from the installer menu) - Now possible to fit large models onto GPUs w/o external libraries: pipeline and model parallelism - determinants & eigenvalues via torch.linalg w/o switching to NumPy

If you can, take your laptop and sit on a porch...then you can enjoy The StatQuest Introduction to PyTorch!!! #BAM youtu.be/FHdlXe1bSe4 Grid.ai @PyTorchLightnin

The greatest Unsolved problems in Modern Physics. A Visual Thread 👇

One of the finest commencement speech I've ever heard. Roger Federer's cited life experiences are so much relatable. I guess, now I can only laugh back and 😉 at them. Never give up! #rogerfederer #motivationquotes

BREAKING: MIT just completed the first brain scan study of ChatGPT users & the results are terrifying. Turns out, AI isn't making us more productive. It's making us cognitively bankrupt. Here's what 4 months of data revealed: (hint: we've been measuring productivity all wrong)

And yet it remains popular in all sorts of disciplines #complexity #dynamicalsystems