Day 52: TensorFlow 1.9

☆ incredible visuals by @SCMPgraphics
"How the Thai cave rescue mission was developed"
https://t.co/jhfT5BQd3p h/t @flowingdata #ThaiCaveRescue pic.twitter.com/gY8bMF0EPJ

— Mara Averick (@dataandme) July 10, 2018

In the spirit of Twitting, I recently completed a paper that summarizes the #BookOfWhy in seven words, each standing for a principle or a tool:https://t.co/kGo69bX109
Once you acquire these seven tools you would qualify as top Causal Inference Expert - a champion of commonsense.

— Judea Pearl (@yudapearl) July 11, 2018

TensorFlow 1.9 is out, with lots of important new features, and a new getting-started guide based on tf.keras and eager execution.

Whether or not you're a TF user, do check it out: TF has made immense progress recently. Big steps towards the future of ML https://t.co/LAGOoRnfPi

— François Chollet (@fchollet) July 11, 2018

Updated TensorFlow Keras documentation is a treat to read! Keras’s inspiring user empathetic API design and documentaion(models, layers, callbacks etc) + Tensorflow’s super charged new capabilities(eager execution, seamless MultiGPU, Dataset) https://t.co/YA3vbmF84z https://t.co/3r1Ekj6CrB

— Sravya Tirukkovalur (@sravsatuluri) July 11, 2018

Check out this end-to-end example of generating Shakespeare-like text using tf.keras + eager → https://t.co/icSa6DQ0GQ pic.twitter.com/aixjtlHtTS

— TensorFlow (@TensorFlow) July 10, 2018

New post with @JasonPunyon! Classifying comments on @StackOverflow ⚖️https://t.co/jvGSFIAkCW pic.twitter.com/Mrs2DjdI9H

— Julia Silge (@juliasilge) July 10, 2018

An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution, from UberAI. Adding CPPN-style coordinate information actually complements and improves spatial-invariant property of ConvNets. Includes a TF implementation in the Appendix. https://t.co/qDfuJc7JuL pic.twitter.com/y7FEQkF7No

— hardmaru (@hardmaru) July 11, 2018

Here's a fun 8-minute YouTube video explaining their CoordConv technique. Wish more authors did this! https://t.co/HrnVbjXvQy

— hardmaru (@hardmaru) July 11, 2018

Our latest work is out!
Representation Learning with Contrastive Predictive Coding (CPC).

Autoregressive modeling meets contrastive losses in the latent space.
Learn useful representations in an unsupervised way.
-> On Audio, Vision, NLP and RL.

Arxiv: https://t.co/HN0zChI9he pic.twitter.com/3Tnpqt9N0v

— Aäron van den Oord (@avdnoord) July 11, 2018

Smoothing the max operator in a dynamic program recursion induces a random walk on the computational graph. The expected path on that walk can be computed efficiently by backpropagation, which converges to backtracking as smoothing vanishes. https://t.co/AjyGsez1B1 pic.twitter.com/RZELfrmRqn

— Mathieu Blondel (@mblondel_ml) July 10, 2018

Self-Guessing Mapper: Extreme Generalization with Topological Modeling & AIThttps://t.co/jo2bWpnLzs

Extends deep learners to work with zero data/increases extrapolation power

- Solve @GaryMarcus Challenge
- Learn Fizz Buzz with MLP
- Exploit Iris dataset
- Manifold Imputation pic.twitter.com/LnfiRFLhht

— MLWave (@MLWave) July 10, 2018

What are radiological deep learning models actually learning? When they're looking for pneumonia outside the lungs, we should all be asking this question.https://t.co/qjPxDEIpFa

— John Zech (@johnrzech) July 8, 2018

Research published in Nature describes an artificial neural network made out of DNA that can solve a classic machine learning problem: correctly identifying handwritten numbers. The work is a step towards programming AI into synthetic biomolecular circuits https://t.co/fxileiitLi pic.twitter.com/4RzKdAPZFj

— nature (@nature) July 9, 2018

https://t.co/Q6U9aSkkaG

I am happy to announce that @AlBuslaev @creaf and I are open-sourcing a library for Image Augmentations.

It is fast. It is diverse. Supports classification, segmentation, detection out of the box. Was used to win a number of #DeepLearning competitions. pic.twitter.com/o1fNk1JN0j

— Vladimir Iglovikov (@viglovikov) July 10, 2018

Nice tool !
"New package 'packagefinder' - Search for packages from the #RStats console" 🔎🔍https://t.co/8wxncExzyo pic.twitter.com/aqIbhH2YRg

— Colin Fay (@_ColinFay) July 10, 2018

Little Python command-line trick that I use very often:

$ python -m http.server

Launch a simple filesystem-backed webserver in one line!

— Jake VanderPlas (@jakevdp) July 10, 2018

As someone who asks for odds ratios *and* relative risk at the vet 🐶, I 🖤 this post…
"How the odds ratio confounds: a brief study in a few colorful figures" by Keith Goldfeld https://t.co/d3Y0wbmxmE #rstats #statistics pic.twitter.com/WOdbmeJxBb

— Mara Averick (@dataandme) July 10, 2018

Using the jupyter-matplotlib project, you can enable interactivity of @matplotlib figures in the @ProjectJupyter notebook and #jupyterlab. To get started: `pip install ipympl`.https://t.co/42cnMNYwUU pic.twitter.com/cdPkxuohf7

— Sylvain Corlay (@SylvainCorlay) July 9, 2018

TIL Jupyter Notebooks can now share kernels within Jupyter Lab. Makes it easy to share state between notebooks. #Python #SciPy2018https://t.co/EL1yHIhbXP pic.twitter.com/zERcRSBA9e

— Randy Olson (@randal_olson) July 10, 2018

Edward - Probabilistic Modeling Made Easy by Maja Rudolph - https://t.co/yQkReJomvU. An introducion to the tensorflow & Edward basics that are necessary to look at a few modeling examples. Examples cover how to fit a Bayesian neural network & an embedding model to real data.

— Python Software (@ThePSF) July 10, 2018

Excellent post on generalisation of Bayesian methods by @sebnowozin Do Bayesian Overfit? (you would never guess the answer;))https://t.co/9CEpDShhSk

— Ferenc Huszár (@fhuszar) July 11, 2018

How many random seeds are needed to compare #DeepRL algorithms?

Our new tutorial to address this key issue of #reproducibility in #reinforcementlearning

PDF: https://t.co/7eHOzhtLuC

Code: https://t.co/0CRRM8RYYr

Blog: https://t.co/rYWM5zPYZB#machinelearning #neuralnetworks

— Pierre-Yves Oudeyer (@pyoudeyer) July 6, 2018

"We need to stop teaching abstinence-only statistics, telling students they can only do it if they are in a committed relationship with a statistician. But all their friends are out there doing statistics and having a great time.

We need to teach safe stats."

- @hadleywickham

— Kelly Bodwin (@KellyBodwin) July 10, 2018

Here’s what @dpatil has been hinting at: Doing Good Data Science—first of a series on data science and ethics. By DJ, @hmason, & @mikeloukides https://t.co/jNQOiQ0X4t

— Mike Loukides (@mikeloukides) July 10, 2018

Look at these actually decent machine learning icons! https://t.co/UK4C8BNL4f pic.twitter.com/z7bPYuUGgU

— Rachael Tatman @ ICML (@rctatman) July 2, 2018

We've made #pytorch code for our ICLR paper https://t.co/6nplgeGOWF with @MILAMontreal and @MSRMontreal on learning sentence representations public - https://t.co/bXoSaYVyjL. Repo includes pre-trained models.

— Sandeep (@sandeep1337) July 10, 2018

To this day when I see the runif() function I first read it as "run if" and not "r unif". #rstats

— Sharon Machlis (@sharon000) July 10, 2018

its really not that hard `which python` https://t.co/2pqUZP4RWM

— Mahdi Yusuf (@myusuf3) July 10, 2018