Day 28: Highlight ggplot2 with Predicates

Solving the Rubik's Cube Without Human Data https://t.co/TIAxBbO54O

— Nando de Freitas (@NandoDF) June 16, 2018

Interpretability in neural nets is amazingly hard. Here's our stab at it:https://t.co/XJGVAhnKbD pic.twitter.com/On8DoqdIlH

— Adam Kosiorek (@arkosiorek) June 15, 2018

Tools for PyTorch https://t.co/swXLBCQQuv #pytorch #deeplearning #machinelearning #ml #ai #neuralnetworks

— PyTorch Best Practices (@PyTorchPractice) June 16, 2018

gghighlight📦 is completely re-designed!📈🔦 Feedbacks are welcome! #rstats

Re-introduction to gghighlight: Highlight ggplot2 with Predicates | Wannabe Rstats-fu https://t.co/1wOzGv203C pic.twitter.com/F6aZPe7VFe

— Hiroaki Yutani (@yutannihilat_en) June 16, 2018

learn deep learning: https://t.co/MhwV380FWg
computational linear algebra: https://t.co/CY7Gu8IXmZ
data science career advice: https://t.co/jKNchZdtA8
myths about AI: https://t.co/DPlLoOLF63
biased AI: https://t.co/3qOtervksj
debunking the pipeline myth: https://t.co/qIW64dWkUg

— Rachel Thomas (@math_rachel) June 16, 2018

This is the best walk thru of our NLP transfer learning paper yet! 😀
Thanks @yashuseth for taking the time to write this up and share it. https://t.co/3BrwIP5Goa

— Jeremy Howard (@jeremyphoward) June 17, 2018

“In-Browser object detection using YOLO and TensorFlow.js” by @PiotrSkalski92 https://t.co/Ny32S9ECrv #yolo #tensorflow #deeplearning #machinelearning #ml #ai #neuralnetworks

— TensorFlow Practices (@TFBestPractices) June 16, 2018

annotations very helpful, might add more, use slightly smaller type. Annnotations can also be images, sparklines, or local zooms https://t.co/IiVrFSz3bx

— Edward Tufte (@EdwardTufte) June 16, 2018

In Python 3, identifiers can contain Unicode letters & digits. This can make equations easier to read: pic.twitter.com/0WLQ6wUd8R

— Aurélien Geron (@aureliengeron) June 16, 2018

Still a fave 📽 deck:
"Introducing Monte Carlo Methods w/ R"https://t.co/WpQL4DDi6t #rstats #mcmc pic.twitter.com/gZeWDEoG7R

— Mara Averick (@dataandme) June 16, 2018

💪 Pipeline mgmt, for reproducible research…
👨‍🔬 "The prequel to [@wmlandau's] drake R package" https://t.co/mVn0HNecqu via @rOpenSci #rstats pic.twitter.com/ffV1I0H02z

— Mara Averick (@dataandme) June 17, 2018

ICYMI, 😭 missed @apreshill's talk? 💖 slides are 💥!
📽 "Big Magic with R: Creative Learning Beyond Fear"
https://t.co/QJQMIXG0Mc #rstats #cascadiarconf pic.twitter.com/Nh5bGTLY3x

— Mara Averick (@dataandme) June 16, 2018

A school in China scans students in the classroom every 30 seconds. pic.twitter.com/YQ2Eql0mni

— hardmaru (@hardmaru) June 17, 2018

Artificial intelligence in teaching https://t.co/oC5MKkq4zo

— Nando de Freitas (@NandoDF) June 16, 2018

“Human cognition and behavior is the product of the interaction of genetic and cultural evolution. Gene-culture co-evolution has allowed us to adapt to highly diverse ecologies and to produce cultural adaptations and innovations.” https://t.co/HRTEr1oIZb

— hardmaru (@hardmaru) June 16, 2018

“We find that a graphical correction decreases misperceptions... more than an equivalent text correction" Very interesting article & study: https://t.co/SzZPGMC56M via @jdschramm #dataviz

— Cole Knaflic (@storywithdata) June 16, 2018