Introduction to the Hands-on Exercise

About this exercise:

In this lesson, we'd like to take you on a bit of a side journey. It's a fun exercise and we hope
you enjoy it as much as we enjoyed putting this short exercise together for you.
Image Classification with IBM Watson Visual Recognition:
What better way to understand the applications of AI than to try it out yourself? You'll be
uploading images and seeing how IBM Watson identifies the various objects in your images.

To learn how to classify your own images, continue on to the next reading!

Hands on Lab: Classify your images with AI!

Lab overview:
Scenario
IBM Watson Visual Recognition (VR) is a service that uses deep learning algorithms to identify
objects and other content in an image. In this hands-on lab, you will use Watson VR to upload
and classify images.
Note: To complete this exercise, you will create an IBM Cloud account and provision an
instance of the Watson Visual Recognition service. A credit card is NOT required to sign up for
an IBM Cloud Lite account and there is no charge associated in creating a Lite plan instance of
the Watson VR service. Objectives
Objective for Exercise:
 How to access IBM Cloud
 How to add resources to your IBM Cloud account
 How to add services to your IBM Cloud account
 How to create a project in Watson Studio
 How to analyze images using Watson VR
Exercise 1: Create an IBM Cloud Account
Scenario
To access the resources and services that the IBM Cloud provides, you need an IBM Cloud
account.
If you already have an IBM Cloud account, you can skip Tasks 1 and 2 and proceed with Task 3:
Login to you IBM Cloud account.
Task 1: Sign up for IBM Cloud
1. Go to: https://cloud.ibm.com/registration to create a free account on IBM Cloud.
Note: If you already have an IBM Cloud account, Login by going
to https://cloud.ibm.com/login and skip ahead to Exercise 2.
2. Enter your company Email address and a strong Password and then click
the Next button.

3. An email is sent to the address that you signed up with to confirm your email address.
Check your email and copy and paste Verification code. Then click Next.

4. Once your email is successfully verified, enter your First name, Last name,

and Country and click Next.
 5. After your personal information is accepted Click on Create account button.

Exercise 2: Create an instance of Watson Studio service

Scenario
Watson Studio, which is available as a service on IBM Cloud, and a component of the IBM
Cloud Pak for Data, allows you to build and collaborate on AI and Data Science projects. In this
exercise, you will add an instance of the Watson Studio service to your IBM Cloud account.
Task 1: Add Watson Studio as a resource
1. In the IBM Cloud Catalog go to the Watson Studio listing
- cloud.ibm.com/catalog/services/watson-studio
2. On the Watson Studio page, select the region closest to you, verify that the Lite plan is
selected, and then click Create.
 3. Once the Watson Studio instance is successfully created, click on Get Started.

Exercise 3: Create a project

Scenario
A project within Watson Studio allows you to organize resources to work with data. You will
begin by creating an empty project, and then adding the resources and services that you need.
Task 1: Create an empty project
1. On the Watson Studio Welcome page, click Create a project.

2. On the Create a project page, click Create an empty project.

3. On the New project page, enter a Name and Description for your project.

4. You must define storage for your project before you can create it. If you already have an
instance of Cloud Object Storage, you can select it, otherwise under Select storage
service, click Add.
5. On the Cloud Object Storage page, verify that Lite plan is selected, and then click Create.

6. Now on the Create Project page, under Define storage, you may need to click Refresh to
see the newly created instance of Cloud Object Storage (COS).
7. Once an instance of Cloud Object Storage (COS) is listed, click Create to create the
Project.

8. A project should now be created in Watson Studio.

Exercise 4: Add a Watson VR Service instance
Scenario
This project will focus on analyzing images, so you need to add an instance of the Watson
Visual Recognition Service. You will also need to collect some images to analyze.
Task 1: Add the Visual Recognition Service
1. To add services to the project, click Add to project.

2. In the Choose asset type box, click Visual Recognition.

3. In the Associate a service box, click here.

4. Click on New service +. If you already have an instance for Watson Visual Recognition
service, skip to step 8.

5. In the panel that is presented scroll down and click on the Visual Recognition service.

6. On the Visual Recognition page, verify that Lite plan is selected, and then click Create.
 7. Your instance of Visual Recognition service should now be created.

8. Now associate your instance of Visual Recognition service with your Watson Studio
project by selecting the check box to the left of it and clicking on Associate service.

Task 2: Collect a few images to be classified by Watson

Before you begin the next task, you will require a few images to be used for classification by
Watson:
 1. Collect a set of at least 5-10 images. You can use your own images, or download them
from the internet. In case your company does not allow downloads, there may already be
some screensavers or background images on your device.
2. Store the images in an easy to find location on your device.
Task 3: Analyze images with Watson VR
Back in your Visual Recognition project you will now see the built-in image classification
models that IBM Watson provides! Let's try the General model.
1. To analyze your images, on the Models page, under Pre Built Models, in
the General box, click Test.

2. On the General page, click the Test tab.

3. To upload images, on the Test tab, click Browse.

4. Select the images you want to upload and then click Open. You can also drag and drop
images from our desktop or file browser onto the Test tab in Visual Recognition.

5. Once you have uploaded your images, Watson Studio Visual Recognition will tell you
what it thinks it found in your images! Beside each class of object (or color, object type
or other characteristics), it gives you a confidence score (between 0 and 1) showing how
confident it is that it found that particular object or feature in your image (0 for lowest
confidence and 1 for highest confidence).
6. Use the check boxes on the left to filter the images. In this example, only images in
which Watson VR has detected beige color are displayed.

7. Use the Threshold slider to only display images in which Watson VR has at least 90%
confidence of the beige color.
Task 3: Save a screenshot
Note: The screenshot saved in this step will be required as part of the Graded Final Assignment
for those pursuing a certificate for this course. This step is optional for those auditing the course.
1. From Task 2, choose just one of the images that you uploaded. Select an image that does
not have too many classes of objects.
2. Take and save a screenshot in .jpeg or .jpg format including the Watson Visual
Recognition confidence scores (that are indicated below the image). Ensure the labels and
confidence scores below the picture are readable. See the sample screenshot below.
Task 4: Share your results
Follow Rav on Twitter and share some of the funniest and most interesting results you found
with IBM Watson Visual Recognition!

Click hereto share the above Tweet.

Author(s)
Polong Lin
Rav Ahuja
Contributor(s)
Rose Malcolm, Shubham
Changelog
Date Version Changed by Change Description
2020-09-22 3.0 Rav Ahuja Updated lab with new IBM Cloud screenshots and steps
Read-only memory (ROM) is a type of memory that can permanently or semi-permanently hold data.
It is called read-only because it is either impossible or difficult to write to. ROM is also often called
non-volatile memory because any data stored in ROM remains even if the power is turned off. Hence
ROM is the ideal place to store the PC’s start-up instructions—that is, the software that boots the
system.

In fact, ROM is technically a subset of the system’s RAM. In other words, a portion of the system’s
random access memory address space is mapped into one or more ROM chips. This is necessary to
contain the software that enables the PC to boot up; otherwise, the processor would have no program
in memory to execute when it is powered on.

For example, when a PC is turned on, the processor automatically jumps to address FFFF0h,
expecting to find instructions to tell the processor what to do. This location is exactly 16 bytes from
the end of the first megabyte (1024 KB) of RAM space, as well as the end of the ROM itself.

If this location were mapped into regular RAM chips, any data stored there would have disappeared
when the power was previously turned off, and the processor would subsequently find no instructions
to run the next time the power was turned on. By placing a ROM chip at this address, a system start-
up program can be permanently loaded into the ROM and will be available every time the system is
turned on.
 Normally, the system ROM starts at address E0000h or F0000h, which is 128KB or 64KB prior
to the end of the first megabyte. The ROM programs occupy the entire last 128KB of the first
megabyte, including the critical FFFF0h start-up instruction address, which is located 16 bytes from
the end of the BIOS space.

Some motherboard ROM chips are larger, up to 256KB or 512KB in size. The additional code
in these is configured to act as a video card ROM (addresses C0000h–C7FFFh) on motherboards with
built-in video and might even contain additional ROM drivers configured anywhere from C8000h to
DFFFFh to support additional onboard devices, such as SCSI or network adapters.

Figure 5.2 shows a map of the first megabyte of memory in a PC; notice the upper memory areas
reserved for adapter card and motherboard ROM BIOS at the end of the first megabyte.

Some think it is strange that the PC would start executing BIOS instructions 16 bytes from the
end of the ROM, but this design is intentionally built in to Intel’s x86 processors. All the ROM
programmer has to do is place a JMP (jump) instruction at that address that instructs the processor
to jump to the actual beginning of the ROM—in most cases, close to E0000h / F0000h—which is about
128KB / 64KB earlier in the memory map.

It’s like deciding to read every book starting 16 pages from the end and then having all
book publishers agree to place an instruction there to jump back the necessary number of pages to get
to page 1. By setting the processor startup location in this way, Intel enabled the ROM to grow to be
any size, all the while keeping it at the upper end of addresses in the first megabyte of the memory
address space.
Adapter card ROMs are automatically scanned and read by the motherboard ROM during the early
part of the boot process—during the POST. The motherboard ROM scans a special area of RAM
reserved for adapter ROMs (addresses C0000–DFFFFh) looking for 55AAh signature bytes. This area
indicates the start of a ROM.

All adapter ROMs must start with 55AAh; otherwise, the motherboard won’t recognize them. The
third byte indicates the size of the ROM in 512-byte units called paragraphs, and the fourth byte is the
actual start of the driver programs. The size byte is used by the motherboard ROM for testing
purposes. The motherboard ROM adds all the bytes in the ROM and divides the sum by the number
of
bytes. The result should produce a remainder of 100h. Thus, when creating a ROM for an adapter,
the
programmer typically uses a “fill” byte at the end to get the checksum to come out right. Using this
checksum, the motherboard tests each adapter ROM during the POST and flags any that appear to
have been corrupted.
The motherboard ROM automatically runs the programs in any adapter ROMs it finds during the
scan. You see this in most systems when you turn them on, and during the POST you see the video
card BIOS initialize and announce its presence.