VT2018 Apache Mxnet
Introduction
Apache MXNet est un framework open-source de deep learning, utilisé pour former et déployer des réseaux neuronaux profonds. Il est évolutif, permettant une formation rapide aux modèles, et supporte un modèle de programmation flexible et de multiples langages de programmation (comme C++, Python, Julia, Matlab, JavaScript, Go, R, Scala, Perl, et Wolfram Language).
La bibliothèque MXNet est portable et peut évoluer vers plusieurs GPU et plusieurs machines. Le MXNet est soutenu par des fournisseurs publics de cloud computing, dont Amazon Web Services (AWS) et Microsoft Azure. Amazon a choisi le MXNet comme cadre d'apprentissage approfondi de choix chez AWS. Actuellement, le MXNet est soutenu par Intel, Dato, Baidu, Microsoft, Wolfram Research et des institutions de recherche comme Carnegie Mellon, MIT, University of Washington et la Hong Kong University of Science and Technology.
Résumé
Abstract
Apache MXNet is a deep learning open-source framework used to train and deploy deep neural networks. It is scalable, allowing rapid model training, and supports a flexible programming model and multiple programming languages (such as C++, Python, Julia, Matlab, JavaScript, Go, R, Scala, Perl, and Wolfram Language).
The MXNet library is portable and can be upgraded to multiple GPUs and machines. MXNet is supported by public cloud computing providers, including Amazon Web Services (AWS) and Microsoft Azure. Amazon has chosen MXNet as its preferred in-depth learning framework at AWS. Currently, MXNet is supported by Intel, Dato, Baidu, Microsoft, Wolfram Research and research institutions such as Carnegie Mellon, MIT, University of Washington and Hong Kong University of Science and Technology.
Synthèse
Sources
- https://github.com/apache/incubator-mxnet
- https://aws.amazon.com/fr/mxnet/
- https://software.intel.com/en-us/neural-compute-stick/get-started
- https://dzone.com/articles/apache-deep-learning-101-using-apache-mxnet-on-the
- https://github.com/movidius/ncappzoo
- https://software.intel.com/en-us/articles/apache-mxnet-v120-released-with-intel-optimized-cpu-backend
- https://mxnet.apache.org/tutorials/index.html
- https://gluon-crash-course.mxnet.io/ndarray.html