Magnus Hyttsten retweeted
Register for the first ever #TensorFlowWorld!
Join us on October 28–31 for machine and deep learning best practices, use cases, and a firsthand look at TensorFlow 2.0.
Use code "FIRST25" for 25% off Gold, Silver, and Bronze passes (valid July 15 - 19) → goo.gle/2XU9Bup
Pretty awesome given that the implementation is so straightforward.
AI Trained on Old Scientific Papers Makes Discoveries Humans Missed vice.com/en_us/article/neagp… – at Norrtälje, Sverige
On vacation and determined to know everything about Swift before it's over. So happy to have explicit type conversations with my friend compiler. Just covered autoclosures. I must go deeper, with horror-filled delightment. – at London, England
Talk at TensorFlow World!!!
Announcing #TFWorld presented by @OReillyMedia @TensorFlow, happening Oct 28-31! We're looking for speakers—come share your work and help connect the #TensorFlow community with your experience and insights.
Apply by April 23 at tensorflow.world
At office studying Reinforcement Learning, started watching bit.ly/2EHTPxm. Something just happened. My being blends into amazing human nature and science. It all becomes the same. Infinity. Zen.
I love the 'tailgators' Google security puts on campus every so often. Google campus is very accessible and welcome to public but private office space is ... private - don't let tailgators into the office space! Very Googley! I had to reject then entering the building, sorry. – at San Francisco, CA
Much welcome. Don't forget to check out TF hub and tensor2tensor also!!! – at Sunnyvale, CA
Announcing the launch of Dataset Search, a new way for researchers to find the datasets they need, wherever they’re hosted, whether it’s a publisher's site, a digital library, or an author's personal web page. Learn more at goo.gl/BYSouA
Woohoo. I love Swift! teapawt.net/TensorFlow/status/9895…
Swift for TensorFlow is now open source!
It is a new programming model for machine learning combining the performance of graphs and the expressivity of Eager execution, with first-class autodiff built right into the compiler
Learn more on the TF blog ↓ goo.gl/BWvgL2
Magnus Hyttsten retweeted
In Part 3 of our Datasets & Estimators series, we demonstrate how to create a custom #Tensorflow Estimator that mimics DNNClassifier's behavior when solving the Iris problem
Read Part 3 of our blog post here ↓ goo.gl/mnJkY1
Wanna do machine learning with TensorFlow??? Then this is a must! (don't miss part #1 linked in the beginning). Go out there and create some mind-blowing models. And don't forget to tell me all about it! teapawt.net/TensorFlow/status/9326…
Check out the 2nd part of our blog series on #TensorFlow Estimators.
This one describes feature columns in detail ↓ goo.gl/Y4FNBV
Magnus Hyttsten retweeted
TensorFlow 1.4.0rc0 just released: tf.contrib.data is now part of the core as tf.data!
github.com/tensorflow/tensor…
#TensorFlow team is having a great time exhibiting at #githubUniverse 💞 – at South San Francisco, CA
Object detection is HARD, but it's now a walk in the park using #tensorflow. Start your ML engines!!! research.googleblog.com/2017… – at Palo Alto, CA
Having a great time presenting #TensorFlow at the GDG Silicon Valley meetup @gdgsv 🙌 – at Sunnyvale, CA
Excited to be part of the tensorflow team and do demos like these... teapawt.net/tensorflow/status/8656… – at Mountain View, CA
They are in the Stanley Cup finals. What can I say... No words can describe this. Go all the way #sharks.
Magnus Hyttsten retweeted
On May 26 join livestreamed Firebase Q&A with @JamesTamplin @MagnusHyttsten #AskFirebaseLive bit.ly/registerLH – at Santa Clara, CA
















