Overview
MXNet is a flexible, open-source Deep Learning library that is popular for research prototyping and production. Together with the high-level Gluon API interface, Apache MXNet is a powerful alternative to TensorFlow and PyTorch.
This instructor-led, live training (online or onsite) is aimed at data scientists who wish to use Apache MXNet to build and deploy a deep learning model for image recognition.
By the end of this training, participants will be able to:
- Install and configure Apache MXNet and its components.
- Understand MXNet’s architecture and data structures.
- Use Apache MXNet’s low-level and high-level APIs to efficiently build neural networks.
- Build a convolutional neural network for image classification.
Format of the Course
- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.
Course Customization Options
- To request a customized training for this course, please contact us to arrange.
Requirements
- An understanding of machine learning principles
- Python programming experience
Audience
- Data scientists
Course Outline
Introduction
- Apache MXNet vs PyTorch
Deep Learning Principles and the Deep Learning Ecosystem
- Tensors, Multi-layer Perceptron, Convolutional Neural Networks, and Recurrent Neural Networks
- Computer Vision vs Natural Language Processing
Overview of Apache MXNet Features and Architecture
- Apache MXNet Compenents
- Gluon API interface
- Overview of GPUs and model parallelism
- Symbolic and imperative programming
Setup
- Choosing a Deployment Environment (On-Premise, Public Cloud, etc.)
- Installing Apache MXNet
Working with Data
- Reading in Data
- Validating Data
- Manipulating Data
Developing a Deep Learning Model
- Creating a Model
- Training a Model
- Optimizing the Model
Deploying the Model
- Predicting with a Pre-trained Model
- Integrating the Model into an Application
MXNet Security Best Practices
Troubleshooting
Summary and Conclusion