Day 21: Tensorflow 1.9

How data scientists are using Machine learning to assist suicide prevention initiatives. Very cool. I hope ML can make a positive contribution to mental health https://t.co/cpBdUvBN7p

— Nando de Freitas (@NandoDF) June 9, 2018

“TensorFlow 1.9 has Arrived!” by @SagarSharma4244 https://t.co/oqNCR2Xp1b #tensorflow #deeplearning #machinelearning #neuralnetworks #ai

— TensorFlow Practices (@TFBestPractices) June 9, 2018

In @TensorFlow 1.9, it is much easier to use Keras with the Data API: just pass data iterators, specify the number of steps per epoch, and you're good to go! Plus it works in both graph mode and eager mode, kudos to the TF team! pic.twitter.com/URJMgUCgB4

— Aurélien Geron (@aureliengeron) June 9, 2018

Excited to announce the FashionGen Text to Image challenge. The task is to generate fashion items based on design descriptions. Data is annotated by designers, with multiple angles of each item and has over 300,000 samples! @SSENSE @element_ai https://t.co/locRd7Jfxm pic.twitter.com/RquDNr8hQV

— Negar Rostamzadeh (@negar_rz) June 8, 2018

In honor of #ShinyAppreciation, here is an interactive summary of the #NBAFinals using play-by-play data (game 4 data not yet available): https://t.co/panf28N8bL #rstats pic.twitter.com/u9tbQlQcJt

— Scott Davis (@scottyd22) June 9, 2018

New paper: Similarity encoding for learning with dirty categorical variableshttps://t.co/j2JQxy5FcC

Code on https://t.co/9Saecz75MP

Useful for data scientists fighting with dirty data pic.twitter.com/0hmt0FXTfR

— Gael Varoquaux (@GaelVaroquaux) June 8, 2018

“Taming LSTMs: Variable-sized mini-batches and why PyTorch is good for your health” by @_willfalcon https://t.co/wjaPGWNM37 #pytorch #deeplearning #machinelearning #ml #ai

— PyTorch Best Practices (@PyTorchPractice) June 9, 2018

I wrote a short blog post with step-by-step instructions for reproducing the experiments in the world models paper. https://t.co/UmCE5L6vbH pic.twitter.com/LyeGpRgY43

— hardmaru (@hardmaru) June 9, 2018

A static screencap tweetstorm was not sufficient to demonstrate how cool the new @RStudio "Jobs" feature is. So here's a short vid that shows how to run R & Python scripts in the background then immediately use the results in RStudio 1.2 https://t.co/jpTQ3Gy6GK #rstats

— hrbrmstr (@hrbrmstr) June 9, 2018

"Estimating mortality rates in Puerto Rico after hurricane María using newly released official death counts" by @rafalabhttps://t.co/UlqAmtqLT7 #rstats pic.twitter.com/TcyKMiQq8S

— Mara Averick (@dataandme) June 9, 2018

Environmental surveillance with machine learning tools: Detecting ocean poachers and measuring wild life populations. https://t.co/CNfu1x5pkk

— Nando de Freitas (@NandoDF) June 9, 2018

Sports analytics: profile of Sammy Gelfand of Warriors by Marc Stein. https://t.co/r77breClqq

— Edward Tufte (@EdwardTufte) June 9, 2018

I wrote a book on data science for startups using the excellent R packages bookdown by @xieyihui and bigrquery by @hadleywickham https://t.co/hUqgRzZhEN

— Ben Weber (@bgweber) June 9, 2018

Academic disciplines are smart locally, but not all the smart globally. Many fields are about themselves; for example, economics is not all that much about the economy but rather about the field of economics. Something can be correct within the field, but false in the real world.

— Edward Tufte (@EdwardTufte) June 10, 2018

Oh wow, seems like a major pivot. pic.twitter.com/DWfNPZYVay

— Vicki Boykis (@vboykis) June 9, 2018