Some adblockers block the registration window. If you do not see the registration box below, please check your adblocker settings.

General Information

The eScience Center offers a range of free workshops and training courses, open to all researchers affiliated with Dutch research organizations. We organize workshops covering digital skills needed to put reproducible research into practice. These include online collaboration, reproducible code and good programming practices. We also offer more advanced workshops such as GPU Programming, Parallel Programming and Deep Learning.

This is an hands-on introduction to the first steps in Deep Learning, intended for researchers who are familiar with (non-deep) Machine Learning.

The use of Deep Learning has seen a sharp increase of popularity and applicability over the last decade. While Deep Learning can be a useful tool for researchers from a wide range of domains, taking the first steps in the world of Deep Learning can be somewhat intimidating. This introduction aims to cover the basics of Deep Learning in a practical and hands-on manner, so that upon completion, you will be able to train your first neural network and understand what next steps to take to improve the model.

We start with explaining the basic concepts of neural networks, and then go through the different steps of a Deep Learning workflow. Learners will learn how to prepare data for deep learning, how to implement a basic Deep Learning model in Python with Keras, how to monitor and troubleshoot the training process and how to implement different layer types such as convolutional layers.

Where: This training will take place online. The instructors will provide you with the information you will need to connect to this meeting.

When: August 21 - 24, 2023, CET, 09:00 - 13:00 CEST.

Requirements: Participants must have access to a computer with a Mac, Linux, or Windows operating system (not a tablet, Chromebook, etc.) that they have administrative privileges on. They should have a few specific software packages installed (listed below).

Accessibility: We are dedicated to providing a positive and accessible learning environment for all. Please notify the instructors in advance of the workshop if you require any accommodations or if there is anything we can do to make this workshop more accessible to you.

Contact: Please email or training@esciencecenter.nl for more information.

---

Code of Conduct

Participants are expected to follow these guidelines:

• Use welcoming and inclusive language
• Be respectful of different viewpoints and experiences
• Gracefully accept constructive criticism
• Focus on what is best for the community
• Show courtesy and respect towards other community members

Syllabus

Introduction

• Recall the sort of problems for which Deep Learning is a useful tool
• List some of the available tools for Deep Learning
• Recall the steps of a Deep Learning workflow
• Explain why it is important to test the accuracy of Deep Learning system.
• Identify the inputs and outputs of a Deep Learning system.
• Test that we’ve correctly installed the Keras, Seaborn and Sklearn libraries

Classification by a Neural Network using Keras

• Use the deep learning workflow to structure the notebook
• Explore the dataset using pandas and seaborn
• Use one-hot-encoding to prepare data for classification in Keras
• Describe a fully connected layer
• Implement a fully connected layer with Keras
• Use Keras to train a small fully connected network on prepared data
• Plot the loss curve of the training process
• Use a confusion matrix to measure the trained networks’ performance on a test set

Monitor the training process

• Explain the importance of splitting the training data
• Use the data splits to plot the training process
• Set the training goal for your deep neural network
• Measure the performance of your deep neural network
• Interpret the training plots to recognize overfitting
• Implement basic strategies to prevent overfitting

Networks are like onions

• Understand why convolutional and pooling layers are useful for image data
• Use normalization as preparation step for Deep Learning
• Implement a convolutional neural network on an image dataset
• Plot and interpret the training process
• Do a visual inspection of the results of your neural network
• Understand strategies to improve your model based on the plots

Schedule

Day 1

• Introduction
  • Recall the sort of problems for which Deep Learning is a useful tool
  • List some of the available tools for Deep Learning
  • Recall the steps of a Deep Learning workflow
  • Explain why it is important to test the accuracy of Deep Learning system.
  • Identify the inputs and outputs of a Deep Learning system.
  • Test that we’ve correctly installed the Keras, Seaborn and Sklearn libraries

Day 2

• Classification by a Neural Network using Keras
  • Use the deep learning workflow to structure the notebook
  • Explore the dataset using pandas and seaborn
  • Use one-hot-encoding to prepare data for classification in Keras
  • Describe a fully connected layer
  • Implement a fully connected layer with Keras
  • Use Keras to train a small fully connected network on prepared data
  • Plot the loss curve of the training process
  • Use a confusion matrix to measure the trained networks’ performance on a test set

Day 3

• Monitor the training process
  • Explain the importance of splitting the training data
  • Use the data splits to plot the training process
  • Set the training goal for your deep neural network
  • Measure the performance of your deep neural network
  • Interpret the training plots to recognize overfitting
  • Implement basic strategies to prevent overfitting

Day 4

• Networks are like onions
  • Understand why convolutional and pooling layers are useful for image data
  • Use normalization as preparation step for Deep Learning
  • Implement a convolutional neural network on an image dataset
  • Plot and interpret the training process
  • Do a visual inspection of the results of your neural network
  • Understand strategies to improve your model based on the plots

---

Setup

To participate in this workshop, you will need access to software as described below. In addition, you will need an up-to-date web browser.

We maintain a list of common issues that occur during installation as a reference for instructors that may be useful on the Configuration Problems and Solutions wiki page.

Software setup

Please follow these setup instructions in preparation for the workshop.

Make sure to download the required datasets as well:

• Weather dataset
• BBQ labels

Install the videoconferencing client

If you haven't used Zoom before, go to the official website to download and install the Zoom client for your computer.

Set up your workspace

Like other Carpentries workshops, you will be learning by "coding along" with the Instructors. To do this, you will need to have both the window for the tool you will be learning about (a terminal, RStudio, your web browser, etc..) and the window for the Zoom video conference client open. In order to see both at once, we recommend using one of the following set up options:

• Two monitors: If you have two monitors, plan to have the tool you are learing up on one monitor and the video conferencing software on the other.
• Two devices: If you don't have two monitors, do you have another device (tablet, smartphone) with a medium to large sized screen? If so, try using the smaller device as your video conference connection and your larger device (laptop or desktop) to follow along with the tool you will be learning about.
• Divide your screen: If you only have one device and one screen, practice having two windows (the video conference program and one of the tools you will be using at the workshop) open together. How can you best fit both on your screen? Will it work better for you to toggle between them using a keyboard shortcut? Try it out in advance to decide what will work best for you.

This blog post includes detailed information on how to set up your screen to follow along during the workshop.