🎥 Machine Learning Live: Let’s Build a Taxi Fare Predictor
👁 1 раз ⏳ 2743 сек.
Bjoern Rost
Principal Consultant
The Pythian Group Inc.

Regardless of what you’ve seen or heard, machine learning is not that complicated, and simple but valuable use cases that extend traditional analytical capabilities are easily unlocked. In this session, the presenters build a prediction model for taxi fares based on existing public data. Watch as they explore the dataset, derive estimates with basic analytics, and then train a TensorFlow model with several iterations to outperform those estimates. You

🎥 Real Talk with Instagram Data Scientist
👁 4 раз ⏳ 696 сек.
Talking data science with Mansha, a data scientist at Instagram. Previously at Blue Apron, Ernst & Young and UPenn. Want to learn data science with a job guarantee? Check out Springboard’s data science career track: https://www.springboard.com/workshops/data-science-career-track/?utm_source=youtube&utm_campaign=youtube-mansha-dsc&utm_medium=video&utm_term=mansha

0:21 What is data science?
0:42 How did you become a data scientist?
2:14 What does your day to day look like?
3:03 What is your favorite and leas

​A Case Study: Exploiting Neural Machine Translation to Translate CUDA to OpenCL

🔗 A Case Study: Exploiting Neural Machine Translation to Translate CUDA to OpenCL
The sequence-to-sequence (seq2seq) model for neural machine translation has significantly improved the accuracy of language translation. There have been new efforts to use this seq2seq model for program language translation or program comparisons. In this work, we present the detailed steps of using a seq2seq model to translate CUDA programs to OpenCL programs, which both have very similar programming styles. Our work shows (i) a training input set generation method, (ii) pre/post processing, and (iii) a case study using Polybench-gpu-1.0, NVIDIA SDK, and Rodinia benchmarks.

​Pebble’s former Head of Data talks about strategies for breaking into data science

🔗 Pebble’s former Head of Data talks about strategies for breaking into data science
Susan Holcomb has seen it all.

​Extreme Event Forecasting with LSTM Autoencoders

🔗 Extreme Event Forecasting with LSTM Autoencoders
Improve forecasting performance developing a strong Neural Network architecture

youtu.be/p1b5aiTrGzY
arxiv.org/abs/1905.08233

🎥 Few-Shot Adversarial Learning of Realistic Neural Talking Head Models
👁 1 раз ⏳ 334 сек.
Paper:
https://arxiv.org/abs/1905.08233v1

Authors:
Egor Zakharov, Aliaksandra Shysheya, Egor Burkov, Victor Lempitsky

Abstract:
Several recent works have shown how highly realistic human head images can be obtained by training convolutional neural networks to generate them. In order to create a personalized talking head model, these works require training on a large dataset of images of a single person. However, in many practical scenarios, such personalized talking head models need to be learned from a f

🎥 Создание системы рекомендаций при написании текстов в специфичной доменной области
👁 1 раз ⏳ 2119 сек.
Очень часто при написании научных статей людям приходится сталкиваться с различными языковыми проблемами. Среди них можно выделить и корректность текста с точки зрения используемых языковых конструкций, и необходимость соответствия используемой лексики и стиля текста доменной области, в которой проводятся исследования. Отсюда возникла идея создания системы, которая умела бы если не исправлять написанный текст, то по крайней мере давать возможные рекомендации по его корректировке на основе существующих текст

​Supervisely: end-to-end web-platform for Deep Learning and Computer Vision

Наш телеграм канал - tglink.me/ai_machinelearning_big_data
https://medium.com/deep-systems/supervisely-end-to-end-web-platform-for-deep-learning-and-computer-vision-eb8d239826d6?fbclid=IwAR1IAMJHDjLMIJOkj2GwoRzTsOkZmFm4LF62MedHXkrXmigzwr_m95ITIrU

🔗 🔥Supervisely: end-to-end web-platform for Deep Learning and Computer Vision
The shortest path to AI product in Computer Vision: label images, manipulate data, train & deploy neural networks within end-to-end…

​Top 5 BI Tools that You must use for Data Visualization

🔗 Top 5 BI Tools that You must use for Data Visualization
Business Intelligence (BI) — The topic of discussion in the business domain since quite a while now. Nearly all kinds of businesses are…

Multithreading In Python | Python Multithreading Tutorial | Python Tutorial
https://www.youtube.com/watch?v=JnFfp81VbOs

🎥 Multithreading In Python | Python Multithreading Tutorial | Python Tutorial For Beginners | Edureka
👁 1 раз ⏳ 1428 сек.
** Python Certification Training: https://www.edureka.co/python **
This Edureka Live video on 'Multithreading in Python'' will help you understand the concept of threading in python. Below are the topics covered in this live video:

What is multitasking in Python?
Types of multitasking
What is a thread?
How to achieve multithreading in Python?
When to use multithreading?
How to create threads in Python?
Advantages of multithreading

Python Tutorial Playlist: https://goo.gl/WsBpKe
Blog Series: http://bit.

🎥 Kaggle Reading Group: Generating Long Sequences with Sparse Transformers | Kaggle
👁 4 раз ⏳ 3625 сек.
Join Kaggle Data Scientist Rachael as she reads through an NLP paper! Today's paper is "Generating Long Sequences with Sparse Transformers" (Child et al, unpublished). You can find a copy here: https://arxiv.org/pdf/1904.10509.pdf

SUBSCRIBE: https://www.youtube.com/c/kaggle?sub_...

About Kaggle:
Kaggle is the world's largest community of data scientists. Join us to compete, collaborate, learn, and do your data science work. Kaggle's platform is the fastest way to get started on a new data science project.

🎥 Lesson 8. Convolutional Neural Networks (practice 1)
👁 1 раз ⏳ 1281 сек.
Lecturer: Gregory Leleytner (ML & DL Researcher at PSAMI MIPT)

Materials: http://bit.ly/2HHuanD

---

About Deep Learning School at PSAMI MIPT

Official website: https://www.dlschool.org
Github-repo: https://github.com/DLSchool/dlschool_english

About PSAMI MIPT

Official website: https://mipt.ru/english/edu/phystechschools/psami
Bachelor's program at PASMI MIPT: http://cs-mipt.ru
Online Master's program at PASMI MIPT: https://mipt.ru/education/departments/fpmi/master/contemporary-combinatoric
Two-year

🎥 TWiML x Fast ai v3 Deep Learning Part 2 Study Group - Lesson 15 - Spring 2019 1080p
👁 2 раз ⏳ 4053 сек.
**SUBSCRIBE AND TURN ON NOTIFICATIONS** **twimlai.com**

This video is a recap of our TWiML Online Study Group.

In this session, we had a mini presentation on "Imagenet-Trained CNNs are Biased Towards Texture; Increasing Shape Bias Improves Accuracy and Robustness" and discussion.

It’s not too late to join the study group. Just follow these simple steps:

1. Head over to twimlai.com/meetup, and sign up for the programs you're interested in, including either of the Fast.ai study groups or our Monthly Meetu

​A Gentle Introduction to Deep Learning : Part 4

🔗 A Gentle Introduction to Deep Learning : Part 4
Probability and Probability Distributions

​Predicting Airbnb prices with machine learning and deep learning

🔗 Predicting Airbnb prices with machine learning and deep learning
Experimentation with XGBoost and tuning neural networks

​OpenAI GPT-2 writes alternate endings for Game of Thrones

🔗 OpenAI GPT-2 writes alternate endings for Game of Thrones
I trained the GPT-2 language model on GRRM’s book series “A Song of Ice and Fire” and let it complete the HBO show’s storyline. Can it do…

​Building Stock Selection into an Artificial Intelligence Framework

🔗 Building Stock Selection into an Artificial Intelligence Framework
A systematic approach to creating an AI driven framework that aligns with your trading goals and investing strategy.

​Wolfram Engine теперь открыт для разработчиков (перевод)

21 мая 2019 Wolfram Researh объявили о том, что они дали доступ к Wolfram Engine для всех разработчиков софта. Вы можете скачать его и использовать в своих некоммерческих проектах по ссылке

Свободный Wolfram Engine для разработчиков дает им возможность использовать Wolfram Language в любом стеке разработки. Wolfram Language, который доступен в виде песочницы — это это мультипарадигмальный вычислительный язык, лежащий в основе самых известных продуктов Wolfram: Mathematica и Wolfram Alpha. Бесплатный Wolfram Engine также имеет полный доступ к базе знаний Wolfram и ее предварительно подготовленным нейронным сетям. Но для его использования вам необходимо оформить бесплатную подписку на Wolfram Cloud.
https://habr.com/ru/post/453074/

🔗 Wolfram Engine теперь открыт для разработчиков (перевод)
21 мая 2019 Wolfram Researh объявили о том, что они дали доступ к Wolfram Engine для всех разработчиков софта. Вы можете скачать его и использовать в своих неком...

​Neural Talking Head Models
https://arxiv.org/abs/1905.08233

🔗 Few-Shot Adversarial Learning of Realistic Neural Talking Head Models
Several recent works have shown how highly realistic human head images can be obtained by training convolutional neural networks to generate them. In order to create a personalized talking head model, these works require training on a large dataset of images of a single person. However, in many practical scenarios, such personalized talking head models need to be learned from a few image views of a person, potentially even a single image. Here, we present a system with such few-shot capability. It performs lengthy meta-learning on a large dataset of videos, and after that is able to frame few- and one-shot learning of neural talking head models of previously unseen people as adversarial training problems with high capacity generators and discriminators. Crucially, the system is able to initialize the parameters of both the generator and the discriminator in a person-specific way, so that training can be based on just a few images and done quickly, despite the need to tune tens of millions of parameters. We sh