# Scalable Deep Learning with Apache MXNet

- Institution: CERN
- Department: EP-IT Data science seminars
- Speaker: Nathalie Rauschmayr
- Year: 2019
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3008381)
- Licence: [CERN](http://copyright.web.cern.ch)
- Who can watch: Tweens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

MXNet is a fully featured, flexibly programmable, and ultra-scalable deep learning framework supporting state of the art deep learning models. It provides both low-level-control and high-level APIs that allow developers to mix imperative and symbolic programming models and to code in their language of choice (including Python, Scala, Java, C++ and R). Besides its computational and memory efficiency, MXNet is lightweight and portable and can run on various systems including edge devices. It supports distributed training on multi-GPUs across multiple hosts and achieves high scalability.

MXNet has its roots in academia and came about through the collaboration and contributions of researchers at several top universities. It originated at Carnegie Mellon University and the University of Washington and is now developed by collaborators from multiple universities and many companies including AWS.

This seminar will give a overview about how MXNet started and an overview of its core capabilities. In addition, the seminar will demonstrate advanced features such as distributed training and deployment on edge devices

.

0:00:00 Slide 1

0:01:10 Slide 2

0:01:48 Slide 3

0:01:52 Slide 4

0:03:40 Slide 5

0:03:44 Slide 6

0:04:47 Slide 7

0:05:08 Slide 8

0:05:33 Slide 9

0:06:29 Slide 10

0:06:33 Slide 11

0:07:05 Slide 12

0:07:09 Slide 13

0:07:48 Slide 14

0:08:03 Slide 15

0:08:11 Slide 16

0:08:41 Slide 17

0:09:29 Slide 18

0:09:55 Slide 19

0:10:16 Slide 20

0:10:29 Slide 21

0:10:34 Slide 22

0:10:42 Slide 23

0:10:47 Slide 24

0:11:10 Slide 25

0:11:14 Slide 26

0:11:26 Slide 27

0:12:39 Slide 28

0:12:44 Slide 29

0:12:48 Slide 30

0:13:37 Slide 31

0:14:21 Slide 32

0:15:20 Slide 33

0:16:00 Slide 34

0:17:49 Slide 35

0:18:47 Slide 36

0:19:14 Slide 37

0:20:10 Slide 38

0:20:20 Slide 39

0:20:52 Slide 40

0:21:15 Slide 41

0:22:02 Slide 42

0:23:34 Slide 43

0:25:01 Slide 44

0:26:11 Slide 45

0:26:38 Slide 46

0:28:18 Slide 47

0:29:06 Slide 48

0:29:22 Slide…

## Lectures

1. [Scalable Deep Learning with Apache MXNet](https://invisiblenetwork.ai/learn/cern/3008381/847e719c08ea), 37 min (CERN's player)