•  
Leader en solutions logicielles pour scientifiques

Reach the next level with Intel in parallel computing

Hands-on Intel® Software Developer Workshop for Technical Computing and Artificial Intelligence

Don’t miss the opportunity to dive deep into the latest tips and tools from Intel® Software and learn more about Intel’s resources and hardware. Test them out yourself and consult our experts onsite at this year’s hands-on training for High Performance Computing & Artificial Intelligence!

Join us for 2 days of hands-on coding sessions on Parallel Programming, Performance Optimization, Artificial Intelligence, and Machine & Deep Learning.

Explore exciting topics including scientific and technical computing, computer vision, image and pattern recognition, machine learning, optimized deep learning and big data analytics.

Please bring your own – Intel®-based – laptop1. We will provide all required software and technology. Detailed technical requirements will be sent to registered attendees.

You are welcome to attend both days but you can also attend one day only. Please specify at registration! Attendance is free, please register as soon as possible as this time we only have very limited seats. After your registration we will review and will confirm your attendance as soon as possible!

Register

DAY 1 : CODE MODERNIZATION FOR INTEL ARCHITECTURE

In this workshop we will dive into the code modernization framework to achieve the highest performance possible. With the help of examples, use cases and a better usage of the Intel® C/C++ compiler, we pinpoint you to possible inefficiencies, hints and strategies to ensure an application delivers great performance.

AGENDA DAY 1**

08:00 - 09:00 Registration with light breakfast

09:30 - 10:00 Getting deeper into Intel® Software Development Tools

For enterprise to cloud, as well as from high-performance computing (HPC) to AI.

This session will introduce intel tools and the different suites available for writing codes for single or multi-nodes computers as well as analyzing the performance.

10:00 - 11:00 Performance optimization using Intel® Compiler with simulation of a dynamical system of particles simulation code - Part 1

Step by step, the attendee will be able to understand how to modify the code of an N-body simulation code, simulation of a dynamical system of particles under the influence of gravity forces, thru different optimization stages spanning from scalar tuning, vectorization tuning, memory tuning and threading tuning.

The session will cover Intel Compiler.

11:00 - 11:15: Coffee Break

11:15 - 13:00: Performance optimization using Intel® Compiler with simulation of a dynamical system of particles simulation code - Part 2

13:00 - 14:00: Lunch Break

14:00 - 16:00 Performance optimization using Intel® Advisor

Intel Advisor is a powerful tool for tracking down and solving vectorization problems. This demo will introduce Intel Advisor and especially the survey and the trip count analyses and we will be able to track all the performance issues of the previous introduced N-body simulation code. We will explain how to read and interpret Advisor's outputs to improve the vectorization and introduce the roofline analysis

This session will cover Intel Advisor and Roofline Model

16:00 - 16:15 Coffee Break

16:15 - 17:00 Maximizing Performance & Scalability Using Performance Libraries

Why would you spend time on optimizing functions in your source code when others already done the work for you ? Intel’s performance libraries like MKL can help to further increase code performance.

17:00 - 19:00 Networking with drinks and finger-food

______________________________

 

DAY 2 : ENHANCE PERFORMANCE WITH INTEL TOOLS AND PYTHON

During this session we will show how performance analysis tools like Intel® Advisor and Intel® VTune™ Amplifier can be used to investigate issues and guarantee the optimal performance. We will show how to use the Intel® Distribution for Python giving insights for machine learning applications.

AGENDA DAY 2**

08:30 - 09:30 Registration with light breakfast

09:30 - 10:00 Getting deeper into the Intel® AI Software Stack for Developers & Data Scientists

Optimized Software for Deep Learning and Machine Learning.

10:00 - 11:30 Faster Machine Learning and Deep Learning training with optimized Frameworks

This session, after first introducting the Intel® Distribution for Python, will mainly cover and demonstrate Intel’s technical best practices for faster time to solution for developing and deploying Machine Learning and Deep Learning training on Intel CPU based platforms.

The session will cover Intel® Distribution for Python, Intel DAAL Library, NumPy / SciPy

11:30 - 12:00 Coffee Break

12:00 - 13:00 Faster Machine Learning and Deep Learning training with optimized Frameworks (part 2)

In this session, we show you techniques we employed to accelerate Tensorflow using the highly-optimized open-source Intel MKL-DNN library.

13:00 - 14:00 Lunch Break

14:00 - 15:30 Case study: Fraud detection with tensorflow, scikit-Learn and Intel® Data Analytics - Acceleration Library (Intel® DAAL)

Credit card fraud is an issue that retailers and many other businesses face. Machine Learning can be used to mitigate this problem. In this session we’ll use different supervised algorithm to detect fraud with a Kaggle dataset. We’ll use Logistic regression, neural networks, random forest and gradient boosting to predict frauds. We’ll also try unsupervised learning techniques such as K-Means clustering and see how they compare with supervised learning algorithms.

We’ll show how Intel has optimised the underlying libraries (TensorFlow and Scikit-Learn) with the MKL and the DAAL and how much speed we get from those optimisations.

15:30-16:00 Coffee Break

16:00-17:30 Deep Learning inference on Intel® based platforms

In this session we will introduce the Intel® Distribution of OpenVINO™ Toolkit.

Based on convolutional neural networks (CNN), the toolkit extends workloads across Intel® hardware (including accelerators) and maximizes performance with deployment on Intel CPU, GPU, Movidius, and Intel FPGAs.

17:30 Q&A Open discussion

Register

Accéder à mon profil

Veuillez rentrer votre email et mot de passe de connexion pour la plateforme MYRITME. S’il s’agit de votre première connexion à MYRITME, veuillez rentrer un email déjà utilisé auprès de Ritme ou bien vous créer un profil en cliquant sur le lien pour créer son profil.

Je souhaite créer mon profil pour bénéficier de tarifs adaptés

Mot de passe oublié ?

Mot de passe oublié ?

Pour recevoir votre mot de passe, veuillez indiquer votre identifiant de connexion (l'adresse e-mail utilisée lors de la création de votre compte).

Creér mon profil

Si vous êtes particulier ou étudiant, Accédez directement à votre catalogue en ligne en créant votre compte avec ce type de profil

* Pays : * Langue :

Cocher votre profil











 
 
 
J’accepte que le groupe Ritme m’adresse ses newsletters, au plus toutes les 2 semaines. Ces communications concernent les actualités et évènements de Ritme (user meetings, rendez-vous thématiques, etc.)
Les e-mails d’informations produit liés aux logiciels et services que vous aurez achetés vous seront envoyés par défaut. Vous pouvez demander à vous rétracter auprès de notre responsable à la protection des données via l’adresse : dataprotection@ritme.com