Building Deep Learning Models Using Apache MXNet

Apache MXNet is the deep learning framework which has its origins at Amazon Web Services (AWS) and is a powerful alternative to TensorFlow. This course teaches you how to build dynamic and static computation graphs using the Gluon API.
Course info
Level
Beginner
Updated
Sep 17, 2018
Duration
2h 3m
Table of contents
Description
Course info
Level
Beginner
Updated
Sep 17, 2018
Duration
2h 3m
Description

Apache MXNet offers low-level and high-level APIs which is key to efficiently build neural networks. It also allows you to construct static and dynamic graphs in a symbolic manner using the Module API, the Symbol API, or the Gluon API. In this course, Building Deep Learning Models Using Apache MXNet, you'll learn the basic building blocks of building neural networks using NDArrays, the Module API, the Symbol API, as well as the cutting edge Gluon API. First, you'll gain an understanding of the basic architecture of MXNet and how the basic data structure NDArrays work. Next, you'll discover the difference between symbolic and imperative programming and when you would choose to use one over the other. Then, you'll discover the use of optimizers, loss functions, and data iterators in building and executing neural networks. Finally, you'll explore the Gluon API and build a convolutional neural network for image classification and hybridize it in order to execute a static computation graph. By the end of this course, you'll have the confidence to efficiently build and execute neural networks using all of the APIs that Apache MXNet has to offer.

About the author
About the author

A problem solver at heart, Janani has a Masters degree from Stanford and worked for 7+ years at Google. She was one of the original engineers on Google Docs and holds 4 patents for its real-time collaborative editing framework.

More from the author
Building Features from Image Data
Advanced
2h 10m
Aug 13, 2019
Designing a Machine Learning Model
Intermediate
3h 25m
Aug 13, 2019
More courses by Janani Ravi
Section Introduction Transcripts
Section Introduction Transcripts

Course Overview
Hi. My name is Janani Ravi, and welcome to this course on Building Deep Learning Models Using Apache MXNet. A little about myself, I have a Master's degree in electrical engineering from Stanford and have worked at companies such as Microsoft, Google, and Flipkart. At Google, I was one of the first engineers working on real-time collaborative editing in Google Docs, and I hold full patent for its underlying technologies. I currently work on my own startup, Loonycorn, a studio for high- quality video content. In this course, you will learn the basic building blocks of building neural networks using NDArrays, the Module and Symbol API, as well as the cutting-edge Gluon API in MXNet. We start off by understanding the basic architecture of MXNet and how the data structure in NDArrays work. We'll understand the difference between symbolic and imperative programming and when you choose to use one over the other. We'll then use symbolic programming with the Module and Symbol API to build a simple classification model for breast cancer detection. We'll understand the use of optimizers, loss functions, and data iterators in building and executing neural networks. We'll then move onto the Gluon API, which is a high-level abstraction to build neural networks imperatively, as well as symbolically. We'll build a convolutional neural network for image classification and then hybridize this network so we can execute a static computation graph. At the end of this course, you should be comfortable building and executing neural networks using all of the APIs that Apache MXNet has to offer.