3.3 Microsoft CNTK (Cognitive Toolkit) / 3.4 NVIDIA DGX-1

Loading...

Do curso por Yonsei University

Deep Learning for Business

204 classificações

Yonsei University

Deep Learning for Business

204 classificações

Your smartphone, smartwatch, and automobile (if it is a newer model) have AI (Artificial Intelligence) inside serving you every day. In the near future, more advanced “self-learning” capable DL (Deep Learning) and ML (Machine Learning) technology will be used in almost every aspect of your business and industry. So now is the right time to learn what DL and ML is and how to use it in advantage of your company. This course has three parts, where the first part focuses on DL and ML technology based future business strategy including details on new state-of-the-art products/services and open source DL software, which are the future enablers. The second part focuses on the core technologies of DL and ML systems, which include NN (Neural Network), CNN (Convolutional NN), and RNN (Recurrent NN) systems. The third part focuses on four TensorFlow Playground projects, where experience on designing DL NNs can be gained using an easy and fun yet very powerful application called the TensorFlow Playground. This course was designed to help you build business strategies and enable you to conduct technical planning on new DL and ML services and products.

Na lição

Deep Learning Computing Systems & Software

The third module “Deep Learning Computing Systems & Software” focuses on the most significant DL (Deep Learning) and ML (Machine Learning) systems and software. Except for the NVIDIA DGX-1, the introduced DL systems and software in this module are not for sale, and therefore, may not seem to be important for business at first glance. But in reality, the companies that created these systems and software are indeed the true leaders of the future DL and ML business era. Therefore, this module introduces the true state-of-the-art level of DL and ML technology. The first lecture introduces the most popular DL open source software TensorFlow, CNTK (Cognitive Toolkit), Keras, Caffe, Theano, and their characteristics. Due to their popularly, strong influence, and diverse capabilities, the following lectures introduce the details of Google TensorFlow and Microsoft CNTK. Next, NVIDIA’s supercomputer DGX-1, that has fully integrated customized DL hardware and software, is introduced. In the following lectures, the most interesting competition of human versus machine is introduced in the Google AlphaGo lecture, and in the ILSVRC (ImageNet Large Scale Visual Recognition Challenge) lecture, the results of competition between cutting edge DL systems is introduced and the winning performance for each year is compared.

3.3 Microsoft CNTK (Cognitive Toolkit) / 3.4 NVIDIA DGX-13:57

Conheça os instrutores

Jong-Moon Chung
Professor, School of Electrical & Electronic Engineering
Director, Communications & Networking Laboratory

Looking into Microsoft CNTK, the cognitive toolkit.

This was originally introduced as CNTK,

listed out as Computational Network Toolkit by Microsoft Research in 2014.

The open source license became available in April of 2015,

and it was re-branded to the name of Cognitive Toolkit in January of 2016.

This is a deep learning neural network open-source license

toolkit included with easy to learn computational steps based on a directed graph.

Python or C++ can be used for programming development in CNTK.

This was original programming model was based upon configuration scripts.

A simplified definition of CNTK network graph nodes are provided here.

Where for input, FeatureNode.

For training labels, LabelNode.

For evaluation- training evaluation, we have CriterionNode,

and for result evaluation,

we have EvalNode, and for the output, we have OutputNode.

The CNTK characteristics.

Well, optimized runtime system for

deep learning neural network training and testing is included.

Abstract computation graphs are used in deep learning neural network designing.

And brain script are used in defining custom networks, improved configuration scripts.

Looking at it and comparing,

CNTK and TensorFlow have many similarities.

Such as, they are both script driven.

These are the references that I used and I recommend them to you.

Next is NVIDIA DGX-1.

Now, this is a deep learning artificial intelligence supercomputer

with fully integrated hardware and software.

And it was made by Nvidia.

These are the system specifications.

Taking a little bit closer look into the GPUs,

the TFLOPS as in terms of performance in terahflops.

The GPU memory, the CPU as well as the maximum power requirement listed down here.

Now, the specifications show that 170 TFLOPS.

That's 10 to the power 12 multiplied to 170 flops.

Deep learning software includes accurate DNN stack and the NVIDIA

NVLink interconnected with

Pascal-powered NVIDIA Tesla P100 GPU accelerators are included.

In addition, it is worthy to note that

the NVIDIA Pascal engine was designed for computer learning, seeing and simulation.

And that's what's included in here.

In addition, the SDK: The Software Development Kit is included.

The SDK helps development of advanced GPU accelerated applications for clouds,

data centers, workstations and embedded platforms.

DIGITS GPU cheap new training system is included.

And drivers and CUDA is included as well.The SDK also includes GPU accelerated libraries,

debugging and optimization tools,

C and C++ compiler,

and a runtime library.

These are the references that I used and I recommend them to you. Thank you.

Explore nosso catálogo

Registre-se gratuitamente e obtenha recomendações, atualizações e ofertas personalizadas.

O Coursera proporciona acesso universal à melhor educação do mundo fazendo parcerias com as melhores universidades e organizações para oferecer cursos on-line.

© 2018 Coursera Inc. Todos os direitos reservados.

Baixar na App Store

Baixar no Google Play