Another good thing that you should read - BERT
https://github.com/google-research/bert
https://github.com/google-research/bert
GitHub
GitHub - google-research/bert: TensorFlow code and pre-trained models for BERT
TensorFlow code and pre-trained models for BERT. Contribute to google-research/bert development by creating an account on GitHub.
Old thing but still valuable - Multi-Head Attention https://mlexplained.com/2017/12/29/attention-is-all-you-need-explained/
How to understand rats language
https://www.youtube.com/watch?v=25LYVxTUZhM
https://www.youtube.com/watch?v=25LYVxTUZhM
YouTube
This algorithm decodes rat squeaks and could revolutionize animal research
Rats are social and chatty critters, but most of their chatter is above the frequency that human ears can hear. If researchers could listen to what rats are “saying,” it could change countless experiments. This week, we try our hand at recording ultrasonic…
Interesting paper on tone mapping, decolorization and downscaling. thanks 2minutespapers to came out with it.
https://arxiv.org/abs/1707.09482
https://arxiv.org/abs/1707.09482
arXiv.org
Deep Feature Consistent Deep Image Transformations: Downscaling,...
Building on crucial insights into the determining factors of the visual
integrity of an image and the property of deep convolutional neural network
(CNN), we have developed the Deep Feature...
integrity of an image and the property of deep convolutional neural network
(CNN), we have developed the Deep Feature...
Nice project about GAN-painting https://github.com/JoYoungjoo/SC-FEGAN
GitHub
GitHub - run-youngjoo/SC-FEGAN: SC-FEGAN : Face Editing Generative Adversarial Network with User's Sketch and Color (ICCV2019)
SC-FEGAN : Face Editing Generative Adversarial Network with User's Sketch and Color (ICCV2019) - run-youngjoo/SC-FEGAN
IBM Watson Studio giveaway for students(if someone use it)
https://medium.com/ibm-watson/watson-studio-desktop-is-now-free-for-academia-f8b50a118370
https://medium.com/ibm-watson/watson-studio-desktop-is-now-free-for-academia-f8b50a118370
Medium
Watson Studio Desktop is now free for academia
Machine Learning, Data Science, and Predictive Analytics techniques are in strong demand. That’s why since its launch, IBM Watson Studio…
Google opensourced their lib GPipe for training really big networks on many machines
https://ai.googleblog.com/2019/03/introducing-gpipe-open-source-library.html?m=1
https://ai.googleblog.com/2019/03/introducing-gpipe-open-source-library.html?m=1
research.google
Introducing GPipe, an Open Source Library for Efficiently Training Large-scale N
Posted by Yanping Huang, Software Engineer, Google AI Deep neural networks (DNNs) have advanced many machine learning tasks, including speech rec...
Good course for those who wants to learn programming for CUDA
https://developer.nvidia.com/udacity-cs344-intro-parallel-programming?fbclid=IwAR0eN3mzmHiBm2FnuyvwKEiTUfJYAdySxR2eVuH9TJfSq4HhjM-9iaAeSIw
https://developer.nvidia.com/udacity-cs344-intro-parallel-programming?fbclid=IwAR0eN3mzmHiBm2FnuyvwKEiTUfJYAdySxR2eVuH9TJfSq4HhjM-9iaAeSIw
NVIDIA Developer
Udacity CS344: Intro to Parallel Programming
What: Intro to Parallel Programming is a free online course created by NVIDIA and Udacity. In this class you will learn the fundamentals of parallel computing using the CUDA parallel computing platform and programming model.
Google as always not doing any marketing on their new services) But we know where to search. Good library of different tutorials based on Google Colab, just click run and you got it.
https://research.google.com/seedbank/seeds
https://research.google.com/seedbank/seeds
Interesting tutorial on solving recurrent tasks with replacing RNN by Ordinary Differential Equations
https://nbviewer.jupyter.org/github/urtrial/neural_ode/blob/master/Neural%20ODEs.ipynb
https://nbviewer.jupyter.org/github/urtrial/neural_ode/blob/master/Neural%20ODEs.ipynb
nbviewer.org
Notebook on nbviewer
Check out this Jupyter notebook!
TensorFlow Dev Talks Summit 2019
https://www.youtube.com/playlist?list=PLQY2H8rRoyvzoUYI26kHmKSJBedn3SQuB
https://www.youtube.com/playlist?list=PLQY2H8rRoyvzoUYI26kHmKSJBedn3SQuB
# Ukraine Only
Ребята в эту субботу 16го Марта в Киеве пройдет Data Science UA Conference. Это действо покруче всяких ММА поединков, потому как травмы после ММА проходят, а боль от МЛ остается с вами на всю жизнь)
Посему это крутая возможность заявить свои претензии на пояс чемпиона в среднем и тяжелом весе или же просто прийти посмотреть и поболеть за ребят)
Cо списокм докладов можно ознакомиться здесь: https://data-science.com.ua/ua/conferences/conference-6/
Приходите будет жарко)
Я тоже там буду, следуйте за парнем с ирокезом и в килте)
А вот что я делаю https://www.facebook.com/anikishaev/videos/2103942513016618/
Ребята в эту субботу 16го Марта в Киеве пройдет Data Science UA Conference. Это действо покруче всяких ММА поединков, потому как травмы после ММА проходят, а боль от МЛ остается с вами на всю жизнь)
Посему это крутая возможность заявить свои претензии на пояс чемпиона в среднем и тяжелом весе или же просто прийти посмотреть и поболеть за ребят)
Cо списокм докладов можно ознакомиться здесь: https://data-science.com.ua/ua/conferences/conference-6/
Приходите будет жарко)
Я тоже там буду, следуйте за парнем с ирокезом и в килте)
А вот что я делаю https://www.facebook.com/anikishaev/videos/2103942513016618/
New awesome tool from Intell to Accelerate Annotation of Digital Images and Video
https://bit.ly/2NZLVBo
https://bit.ly/2NZLVBo
Intel
New Computer Vision Tool Accelerates Annotation of Digital Images and Video
we are pleased to present a new open source program called Computer Vision Annotation Tool (CVAT, pronounced “si-vi-eɪ-ti”) that accelerates the process of annotating digital images and videos for use in training computer vision algorithms. In this post,…
Fast, small, runs on local device. New Google Speech recognition model
https://bit.ly/2JbrClG
https://bit.ly/2JbrClG
research.google
An All-Neural On-Device Speech Recognizer
Posted by Johan Schalkwyk, Google Fellow, Speech Team In 2012, speech recognition research showed significant accuracy improvements with deep learn...