Live Webinar

GUI-Fying the Deep Learning Workflow in 5 Steps
23.04.2020 | 15:00-16:00(GMT+8)
Online via Webex

About The Event

Experiment Manager-new app in MATLAB r2020a

How do you train a Deep Learning Network? 

R2020a logo
has streamlined the workflow with newly launched Deep Network Designer and Experiment Manager in GUI 

Both apps are a point-and-click tool for creating, modifying, training and hyperparameter tuning the deep neural networks.

Are you still doing Coding?

When the workflow is automated in GUI, the scientists and engineers focus on the least amount of code needed to design their AI products. It has 3 major benefits to real-world software projects : less human error in the codebase, faster development , and more variation. While the first 2 points are obvious, the last is  less appreciated. Variation increases with automation because it’s easier to try things when there is less code to write. Variable is a critical ingredient in the early stages of product discovery.

Deep_Network_Designer - MATLAB r2020a New App

Highlights of this Event

In this webinar, we will explore how these both apps automate your deep learning workflow and gain insight into the hyperparameter for training accurate deep learning models.

If you are using older version R2019a/b, we are happy to let you know there is transfer learning app in MathWorks’s File Exchange to automate your classification transfer learning training and code generation.

We will use real-world examples to demonstrate:

Deep Network Designer:

  • Accessing and managing large sets of images
  • Using visualization to design neural network
  • Analyze and validate the network to ensure that the network architecture is defined correctly, and detect problems before training
  • Train neural network for classification problems and select augmentation options.
  • Monitor training with plots of accuracy, loss, and validation metrics.
  • Generate MATLAB code for building and training networks.

Experiment Manager:

  • Sweep through a range of hyperparameter values to train a deep network.
  • Compare the results of using different data sets to train a network.
  • Test different deep network architectures by reusing the same set of training data on several networks.

The Speaker

Kevin-Chng-01
Kevin Chng
Application Engineer | Developer of Transfer Learning App

Save Your Seat

View Upcoming Events